ENHANCED HEAVY OIL RECOVERY BY EMULSIFICATION WITH INJECTED
NANOPARTICLES
A Thesis
by
ARTURO REY MARTINEZ CEDILLO
Submitted to the Office of Graduate and Professional Studies of Texas A&M University
in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE
Chair of Committee, Robert H. Lane
Committee Members, Maria A. Barrufet Yuefeng Sun Head of Department, Daniel Hill
December 2013
Major Subject: Petroleum Engineering
Copyright 2013 Arturo Rey Martinez Cedillo
ii
ABSTRACT
In-situ oil-in-water emulsion generation, using modified silica hydrophilic
nanoparticles as emulsifier, has been proposed as an enhanced oil recovery process. The
nanoparticles are injected as an aqueous dispersion; its hydrophilic character allows
emulsifying the immobile heavy oil, and transports it out of the reservoir as a low
viscosity fluid. Generating the emulsions in the reservoir was suggested because it offers
numerous advantages. The first advantage is low injectivity pressures due to the low
dispersion viscosity. Also, the size of nanoparticles (5 nm) yields a better emulsion
stability. Furthermore, complex injection facilities are not required, which reduces
operational costs.
In this research, 12 nanoparticle dispersions were created using nanoparticle
concentrations of 0.5, 2.0 and 5.0 wt%, deionized water or brine made with 0.5 wt% of
Sodium Chloride. These dispersions were tested to investigate their ability to generate
oil-in-water emulsions. Emulsion generation experiments included interfacial tension
measurements between heavy oil and nanoparticle dispersions, microscopy analysis to
determine the amount of emulsion generated, and emulsion viscosity measurements.
Results obtained from these experiments indicated that the nanoparticles lead to a
reduction of the interfacial tension of the heavy oil and the dispersion. In addition, the
presence of Sodium Chloride in the dispersion reduced still more of that interfacial
tension, generating the largest amount of emulsions.
iii
Six core flooding experiments were conducted to study the effect of the
nanoparticle dispersion flooding on the final recovery under different settings. Two
types of core plugs with permeabilities of 150 mD and 2,300 mD, and two heavy oils
with viscosities of 600 cP and 3500 cP were combined to establish the original
experiment conditions. Tertiary heavy oil recoveries ranged from 20% to 64 % of OOIP
were obtained. The results throughout these experiments suggest that if the reservoir
conditions (e.g. permeability, porosity and oil viscosity) are adequate, the nanoparticle
dispersion flooding may be a reliable alternative to the thermal recovery processes.
iv
DEDICATION
This thesis is dedicated to my wife, Mirsha Blanco for all the love and support
she has provided me since I met her.
v
ACKNOWLEDGEMENTS
I would like to thank my committee chair, Dr. Robert Lane for his teachings and
supervision, and my committee members, Dr. Maria Barrufet and Dr. Yuefeng Sun for
their guidance and support throughout the course of this research.
I would like to thank the Petroleum Engineering Department at Texas A&M
University for giving me the opportunity to achieve this goal. Thanks also go to my
colleagues, faculty and staff for making my time at Texas A&M University a great
experience.
I would also like to thank my friends Raul Gonzales, Andres Del Busto,
Francisco Tovar, Johannes Alvarez, Juan Lacayo, Gorgonio Fuentes and Jose Moreno
for their valuable assistance and their useful advices.
Finally, thanks to my mother Isabel and my brother Guillermo for their
encouragement, patience and love.
vi
TABLE OF CONTENTS
Page
ABSTRACT ................................................................................................................. ii
DEDICATION ............................................................................................................. iv
ACKNOWLEDGEMENTS ......................................................................................... v
TABLE OF CONTENTS ............................................................................................. vi
LIST OF FIGURES ...................................................................................................... viii
LIST OF TABLES ....................................................................................................... xi
I. INTRODUCTION .................................................................................................... 1
1.1 Significance of the non-thermal EOR process for the heavy oil reservoir exploitation ........................................................................................ 3
1.2 Status of the solid-stabilized emulsions for EOR processes ............................ 6
1.2.1 Emulsions as a drive fluid for EOR processes ........................................ 6 1.2.2 Solid stabilized emulsions flooding ........................................................ 8 1.2.3 Nanoparticle-stabilized emulsion flooding ............................................. 8 1.2.4 Heavy o/w emulsion for pipeline transportation ..................................... 9
1.3 Research objectives .......................................................................................... 10 1.4 Overview of methodology ................................................................................ 10
II. HEAVY OIL: A VAST SOURCE OF ENERGY FOR THE FUTURE ................ 12
2.1 Importance of heavy oil due to the natural depletion of conventional oil ...................................................................................................................... 12 2.1.1 Definition of heavy oil ............................................................................ 12 2.1.2 Classification of heavy oil ....................................................................... 13 2.1.3 Degradation of crude oil to origin heavy oil ........................................... 14 2.1.4 Estimation of global resources ................................................................ 15
2.2 Current status of the technologies of heavy oil recovery processes ................. 16
2.2.1 Primary recovery methods ...................................................................... 18 2.2.2 Thermal enhanced heavy oil recovery methods ...................................... 20
2.2.3 Non-thermal enhanced heavy oil recovery methods ............................... 25
vii
Page
III. EXPERIMENTAL DESCRIPTION ...................................................................... 31
3.1 Experimental materials ..................................................................................... 31 3.1.1 Modified silica hydrophilic nanoparticles ............................................... 31 3.1.2 Heavy crude oil ....................................................................................... 31 3.1.3 Sandstone cores ....................................................................................... 32 3.1.4 Brine solution .......................................................................................... 32
3.2 Emulsion generation and properties bench test ................................................ 32
3.2.1 Emulsification ......................................................................................... 32 3.2.2 Properties measurement .......................................................................... 34
3.3 Core flooding system ....................................................................................... 36 3.4 Experimental procedures .................................................................................. 37
3.4.1 Mechanical emulsion generation ............................................................. 37 3.4.2 Core flood experiment procedures .......................................................... 38
IV. ANALYSIS OF RESULTS ................................................................................... 43
4.1 Emulsion generation ......................................................................................... 43
4.1.1 Nanoparticle dispersions and heavy oil interfacial tension ..................... 46 4.1.2 Microscopic emulsion imaging ............................................................... 49 4.1.3 Emulsion rheology measurements .......................................................... 51 4.1.4 Selection of the nanoparticle dispersion for further EOR
experiments ............................................................................................ 54 4.2 Core flooding experiments ............................................................................... 55
4.2.1 Effects of the ND1 on the in-situ emulsion generation and oil recovery ................................................................................................. 57
4.2.2 Effects of the ND2 on the in-situ emulsion generation and oil recovery ................................................................................................. 76
V. CONCLUSIONS AND RECOMMENDATIONS .................................................. 84
5.1 Conclusions ...................................................................................................... 84 5.2 Future work and recommendations .................................................................. 86
REFERENCES ............................................................................................................. 87
viii
LIST OF FIGURES
Page
Fig. 2.1— Importance of the heavy oil lies in the great volume in-place since it represents 70% of the world total resources .......................................... 16
Fig. 2.2— Heavy oil recovery process may be divided in Primary, Non-Thermal and Thermal methods............................................................................. 18
Fig. 3.1— CAFRAMO RZR50 stirrer .................................................................... 33
Fig. 3.2— Brookfield DV-III Ultra Programmable Rheometer .............................. 35
Fig. 3.3— Interfacial tension System OCA 15Pro .................................................. 35
Fig. 3.4— Actual and schematic core flooding system .......................................... 36
Fig. 3.5— Core permeability to brine was measure at different injection rates and its corresponding pressure drop. ..................................................... 41
Fig. 4.1— Samples of emulsions generated using deionized water and brine with Louisiana heavy oil ................................................................................ 45
Fig. 4.2— Sample of emulsions generated using deionized water and brine with Kern River heavy oil .............................................................................. 46
Fig. 4.3— IFT measurements show the reduction of IFT when the nanoparticle and salt concentration is increased......................................................... 47
Fig. 4.4— IFT between Louisiana heavy oil and the nanoparticle dispersions. A reduction in the IFT is observed according to the nanoparticle and NaCl concentration are increased .......................................................... 48
Fig. 4.5— IFT between Kern River heavy oil and the nanoparticle dispersions. IFT was not measured in the final experiment since the oil drop was not formed .............................................................................................. 49
Fig. 4.6— Microscopic images of emulsions made with Louisiana heavy oil and the nanoparticle dispersions ................................................................... 50
Fig. 4.7— Microscopic images of emulsions made with Kern River heavy oil and the nanoparticle dispersions ............................................................ 50
ix
Page
Fig. 4.8— Increments in nanoparticle concentration leads to an increment in the apparent viscosity; moreover, this viscosity is very low compared with the original heavy oil ..................................................................... 52
Fig. 4.9— Increments in apparent viscosity as a result of add NaCl were more noticeable in sample 16.......................................................................... 53
Fig. 4.10— Buff Berea and Bentheimer sandstone cores used in the flooding experiments ............................................................................................ 55
Fig. 4.11— Oil recovery and pressure profiles along the CF1 ................................. 59
Fig. 4.12— Effluents collected from the CF1 ........................................................... 60
Fig. 4.13— Oil recovery and pressure profiles along the CF2 ................................. 62
Fig. 4.14— Effluents collected from the CF2. .......................................................... 63
Fig. 4.15— Oil recovery and pressure profiles along the CF3 ................................. 65
Fig. 4.16— Effluents collected from the CF3 ........................................................... 66
Fig. 4.17— Oil recovery and pressures profile along the CF4 ................................. 68
Fig. 4.18— Effluents collected from the CF2 ........................................................... 69
Fig. 4.19— Recovery production profile from CF1 and CF2 .................................. 71
Fig. 4.20— Pressure profile from CF1 and CF2 ....................................................... 71
Fig. 4.21— Recovery production profiles of CF3 and CF4 ...................................... 73
Fig. 4.22— Pressure profiles of CF3 and CF4 .......................................................... 73
Fig. 4.23— Effect of oil viscosity in the CF1 and CF3 ............................................ 75
Fig. 4.24— Effect of oil viscosity in the CF2 and CF4 ............................................ 75
Fig. 4.25— Oil recovery and pressure profiles along the core flooding 5 ................ 77
Fig. 4.26— Effluents collected from the CF5 ........................................................... 78
x
Page
Fig. 4.27— Oil recovery and pressure profiles along the CF6. Due to the high rock permeability, the pressure drop remains low when it is compared with the lower rock permeability ........................................................... 80
Fig. 4.28— Effluents collected from the CF6 ........................................................... 81
Fig. 4.29— Comparison of the ND1 and ND2 in Buff Berea sandstone .................. 83
Fig. 4.30— Comparison of the ND1 and ND2 in Bentheimer sandstone ................. 83
xi
LIST OF TABLES
Page
Table 2.1— Classification of crude oil according to API gravity. ............................. 13
Table 2.2— Classification of crude oil according to the World Petroleum Congress 14
Table 3.1— Properties of the two heavy oils used in this research ............................ 31
Table 4.1— Sixteen emulsion generation experiments were conducted to identify the optimum nanoparticle concentration to generate the largest amount of emulsions ............................................................................................ 44
Table 4.2— Rock-fluid properties and original saturation conditions ....................... 56
Table 4.3— Core flooding 1 experiment results ......................................................... 58
Table 4.4— Recovery factors and injection pressures at each injection rate, CF1 .... 60
Table 4.5— Core flooding 2 experiment results ......................................................... 61
Table 4.6— Recovery factors and injection pressures at each injection rate, CF2 .... 63
Table 4.7— Core flooding 3 experiment results ......................................................... 64
Table 4.8— Recovery factors and injection pressures at each injection rate, CF3 .... 66
Table 4.9— Core flooding 4 experiment results ......................................................... 67
Table 4.10— Recovery factors and injection pressures at each injection rate, CF4 ... 69
Table 4.11— Core flooding 5 experiment results ......................................................... 76
Table 4.12— Recovery factors and injection pressures at each injection rate, CF5 .... 78
Table 4.13— Core flooding 6 experiment results ......................................................... 79
Table 4.14— Recovery factors and injection pressures at each injection rate, CF6 .... 81
1
I. INTRODUCTION
Fossil fuels have been the principle source of energy in the world in the last
century. However, due to accelerated population growth and decline of conventional
hydrocarbon production, it is necessary to develop new energy sources to supply the
high demand of countries such as the United States or China. To address this problem,
the oil industry has turned to unconventional resources. Every unconventional reservoir
has its own unique characteristics and challenges. This work will focus on heavy oil
reservoirs since they are a vast source of hydrocarbons. According to the International
Energy Agency (IEA), such reservoirs hold more than 70% of the oil resources.
Moreover, they can be found in every continent, with Canada and Venezuela being the
countries with the largest accumulation of them.
Heavy oil is not as valuable as light oil; in addition to this, it is more difficult to
extract, refine and transport because of its high density and high viscosity. However, oil
companies have realized that the volume of heavy oil in place is very significant, more
than 9 trillion barrels in place (Alboudwarej et al. 2006). Therefore, the oil industry sees
in heavy oil reservoirs an opportunity to generate profit by investing early in methods
and techniques to enhance recovery.
Primary heavy oil recovery methods are being applied widely in Canada and
Venezuela. For example, cold heavy oil production with sand (CHOPS) in Canada
2
currently contributes to more than 500,000 bpd (Dusseault and Baoci 2011). Other
techniques involve very complex well geometry, e.g. the Orinoco heavy oil Belt wells in
Venezuela. Nevertheless, thermal enhanced oil recovery (EOR) methods have
demonstrated to be more efficient since they are able to reduce the in-situ viscosity
making the heavy oil flow through the porous media easier. Nonetheless, there are some
adverse circumstances that restrict their use. Depth of the formation, thickness and
surface weather conditions are some constrains that limit the applicability of the
aforementioned techniques.
Non-thermal methods provide a notable alternative to recover heavy oil. Methods
such as water flooding, polymer flooding and emulsion flooding improve the mobility
ratio between the injected and displaced fluid, along with improving the sweep
efficiency. Similarly, alkali-surfactant flooding increases the displace efficiency by
reducing the interfacial tension between the injected and displaced fluid. All these
methods have been applied to heavy oil reservoirs with different levels of success (Shah
et al. 2010).
Poor conformance is the principal factor of unsuccessful heavy oil waterflooding
projects. Emulsifying the injected water with crude oil is an innovative solution to
increase the viscosity of the water and therefore improving the mobility ratio. An
important aspect of the emulsions is to ensure their stabilization throughout the whole
flooding. Surfactants and solid particles are the most common emulsions stabilizers.
3
However, solid particles seem to be the best option because they may preserve
the emulsion for longer periods of time. In addition, they are less expensive with respect
to surfactants. Smaller particles, in the order of 1 to 100 nanometers, may stabilize and
generate small enough emulsions able to flow through the pore throats and also keep the
stabilization no matter the tortuosity of the rock (Zhang, T. et al. 2010).
Currently, emulsion flooding has been conceived as generating the emulsions at
the surface and then injecting them into the reservoir. In this work, the ability of creating
in-situ oil-in-water (o/w) emulsion with solid nanoparticles as stabilizers that will
maximize heavy oil recovery and reduce costs is proposed and evaluated.
1.1 Significance of the non-thermal EOR process for the heavy oil reservoir
exploitation
The EOR processes are commonly used techniques in the oil industry to extract
remaining hydrocarbons in the reservoir after they have been exploited through primary
recovery methods, that is, using only the reservoir’s own energy or with any artificial lift
system. EOR processes are classified in four main categories according to their action
mechanism: waterflooding, chemical methods, miscible methods and thermal methods.
Based on the reservoir’s characteristics, i.e. type of hydrocarbons, depth and thickness,
and reservoir’s conditions, i.e. temperature, and pressure, EOR processes may be applied
not only to maintain or restore the reservoir pressure but they also improve the
volumetric sweep, and the oil displacement efficiency.
4
Unlike conventional reservoirs, some of the heavy oil reservoirs require EOR
processes early in the development stages of the field because they have special features
such as high viscosity/density, which makes them difficult and at times unviable for
exploitation with primary recovery methods. Different sorts of EOR processes have been
tested in heavy oil reservoirs; however, thermal methods have shown to outperform all
the other methods since their main goal is to reduce the oil in-situ viscosity in order to
improve oil displacement efficiency.
Unfortunately, there are some limitations in the application of thermal methods,
i.e. the steam based methods such as steam flood, steam assisted gravity drainage
(SAGD) and cyclic steam stimulation (CSS) are more effective in shallow and thick
reservoirs to avoid heat loss in the wellbore and in adjacent formations respectively.
Also, if the weather conditions are adverse, as in the North Slope or Alaska, it may
reduce the heat efficiency. Moreover, steam generation requires great amounts of energy
which increases the cost of the incremental barrel.
In response to thermal processes limitations, non-thermal processes have gained
importance to recover heavy oil. Waterflooding projects have been carried out for more
than 50 years in Canada(Miller 2005). Nevertheless, the results have not been favorable
due to the difference in viscosities from the injected water and the formation’s heavy oil.
This yields unfavorable mobility-ratios, leading to low recovery factors (Mai and
Kantzas 2009).
5
As technology, research and development advance, different solutions to improve
poor mobility ratios have been tested. Increasing the viscosity of the water injected
using polymers has shown positive results when it is applied in modest viscosity
reservoirs, less than 100 cP (Fu et al. 2012). However, for more viscous oil, high
concentration of polymer is required that makes difficult their injection and increases the
costs of the process.
Alkali-Surfactant (AS) and Alkali-Surfactant-Polymer (ASP) flooding have also
been tested in heavy oil reservoir. Surfactants reduce the interfacial tension between the
injected fluid and the oil. Alkali is used to create in-situ surfactants when it reacts with
acidic oil (Fu et al. 2012). Hence, if the interfacial tension is reduced, oil and water may
be easily emulsified. Emulsions both water-in-oil (w/o) and oil-in-water (o/w) improve
oil recovery because they modify the fluid’s viscosity, and also improve the mobility
ratio and plugging off high permeability channels (Mai et al. 2009).
Emulsions not only may be generated by using surfactants, but also using solid
particles as emulsifiers. Natural clays, such as kaolinite, bentonite, and treated fumed
silica have been found to be effective solids to stabilize the emulsions. There are two
main concerns regarding solid stabilized emulsions, or Pickering emulsions, for
deployment in porous media; one of them involves the particles size because if it is very
large it might plug the pore throat. The other concern involves emulsion stabilization
because the stabilized emulsions must last during the whole duration of the injection
6
project without breaking. The size problem can be addressed using a smaller particle
size, such as nanoparticles. They are small enough to be able to flow through the pore
throats of most conventional heavy oil reservoirs without almost any restriction.
Emulsion flooding can be seen as a technical and economical alternative to
thermal methods. Emulsion flooding is able to improve the mobility ratio, and it can be
applied to recover viscous oils up to 3000 cP (Kaminsky et al. 2010). Furthermore,
emulsions can be generated by mixing produced oil either in the surface or in-situ, which
reduces costs, and increases the efficiency of the EOR project (Bragg 1999).
1.2 Status of the solid-stabilized emulsions for EOR processes
Solids-stabilized emulsion is not a new concept. Pickering (1907) realized that a
sufficiently smaller insoluble solid might act as an emulsifier between the oil and water
increasing the emulsion stability. “Pickering emulsions” have been studied and used in
different areas such as the cosmetic, food, paint and pharmaceutical industries (Arditty et
al. 2004); however, there are only a few projects related to EOR process.
1.2.1 Emulsions as a drive fluid for EOR processes
Application of o/w and w/o emulsion for EOR processes have not been
extensively studied thus far; nonetheless, there are some projects related to emulsions
flooding in order to both displace the oil in-situ, and reduce the water channeling in
heterogeneous reservoirs.
7
McAuliffe (1973) described a field test where an emulsion injection was carried
out to reduce water fingering in a waterflood project. The project consisted of injecting a
“surface-prepared o/w emulsion” into the Midway-Sunset oil field. The emulsion was
prepared using 70% of produce oil and 30% of fresh water with a minimum volume of
sodium hydroxide as a stabilizer. Despite the fact that the emulsions were
thermodynamically unstable due to the lack of a better stabilizer, the results showed that
emulsions plugged high permeability zones increasing oil recovery and reducing water
oil ratios.
The elaboration of o/w and w/o emulsions can also be made mixing used engine
oil, and brine without any additional chemical emulsifiers as demonstrated by Fu and
Mamora (2010). They stated this idea based on easy accessibility of the used engine oil,
besides proposing a disposal alternative of it. They achieved the stabilization of the
emulsion with the engine oil soot particles, surfactant additives already in the engine oil
and by oil oxidation that created additional surfactants. The major constraint of this
method is that the used engine oil will not always have a unique composition, which in
turn makes each emulsion different. Therefore, it would be impossible to have a standard
procedure to create the same emulsion each time. Additionally the emulsification
process must be carried out at the surface, which greatly increases injection pressure of
the drive fluid.
8
1.2.2 Solid stabilized emulsions flooding
The first attempt to apply solid-stabilized emulsions as an EOR method was
developed by Bragg (1999). The emulsions were created mixing crude oil with brine and
then they were stabilized with solid particles such as natural clays (kaolinite and
bentonite) or treated fumed silica. The average particle size ranged from 1 to 2 microns.
The emulsions were injected into a core saturated with heavy oil of 325 cP and a
temperature of 140 °F. As a result, it was observed that after 1 pore volume of injected
emulsions, the oil recovery was 2.1 to 3.8 times greater than the recovery reached by
waterflooding. Furthermore, the emulsions were stable and showed a favorable mobility
control. Although solid stabilized emulsions provided positives results, Bragg suggested
that smaller particles would increase the emulsion stability.
1.2.3 Nanoparticle-stabilized emulsion flooding
Nanoparticles used as emulsion stabilizers have become more important in EOR
processes, because their high adsorption energy allow the emulsions to last for long
periods of time (Saigal et al. 2010). They are also able to flow through the porous media
due to their small size, which is between 1 to 100 nanometers (Kaminsky et al. 2010;
Zhang, T. et al. 2010).
Zhang, T. et al. (2010) from The University of Texas at Austin tested the ability
of applying w/o and o/w nanoparticle-stabilized emulsions in EOR processes. The
emulsions were elaborated using decane, brine and two different types of silica
9
nanoparticles, hydrophilic and hydrophobic. In order to create o/w emulsions, the
hydrophobic nanoparticles were mixed with the decane and the brine. On the other hand,
w/o emulsions were created with the hydrophobic nanoparticles. This study concluded
that the nanoparticles provide high emulsion stabilization, up to several months, and they
may be used to improve the mobility ratio in an emulsion flooding project.
In 2002 a pilot test was implemented by Exxon Mobil in the Celtic oil field in
Canada where a nanoparticle-stabilized w/o emulsion was injected into a sandstone
reservoir (Kaminsky et al. 2010). The emulsion was created at the on-site facilities
mixing 60 vol% of formation water, 36 vol% of crude oil, 4 vol% of propane and less
than 2 grams of oleophilic fumed silica nanoparticles per liter of emulsion. The injection
layout was built with a central producing well, four injection wells and three observation
wells. Three years into the project, it was concluded that the emulsion flooding was
considerably stable and that it was able to flow through the porous media improving the
heavy oil recovery.
1.2.4 Heavy o/w emulsion for pipeline transportation
Pipeline transportation of heavy oil is a significant issue that several authors have
considered; the high density/viscosity and the high contents of asphaltene and paraffin
leads to pipe obstruction and high pressure drops (Martínez-Palou et al. 2011)). Water
external emulsions offer an optimistic solution because they have a lower viscosity,
reducing pipe friction, and therefore needing less boost energy (Ashrafizadeh and
10
Kamran 2010). A successful example of transportation of o/w emulsions is the
Orimulsion. It was a product conceived and patented by Bitumenes Orinoco S.A.
(BITOR) in Venezuela in order to transport the bitumen produced in the Oricono Belt
more efficiently. Additionally, it was used as a fuel for electric power plants substituting
mineral coal (Gómez-Bueno et al. 1998). Unfortunately, the high level of pollutants
generated by its combustion has made it unattractive to the industry (Oppelt 2001).
1.3 Research objectives
The objective of this work is to evaluate the generation of nanoparticle-stabilized
o/w emulsions in a heavy oil saturated sandstone core, as well as to determine the
additional oil recovered by the emulsions. Hydrophilic nanoparticles of 5 nm in size are
injected as water dispersion into the sandstone core to emulsify the heavy oil. The
hydrophilic nature of the nanoparticle allows the water to trap the residual oil droplets
transporting them out of the core.
1.4 Overview of methodology
The methodology followed in this research has the aim to investigate the
potential of hydrophilic nanoparticles to generate heavy o/w emulsions in order to
develop a new enhanced heavy oil recovery method.
The methodology consisted in testing a number of aqueous dispersions with
different nanoparticle concentrations. The dispersions were mixed with heavy oil
11
utilizing a stirrer to provide enough energy to create the emulsions. Once the dispersions
were tested, the one that emulsified the largest amount of heavy oil was selected as the
injection fluid for the later core flooding experiments.
Core flooding tests were performed to prove the in situ emulsion generation by
the injection of nanoparticle dispersion, and to verify that nanoparticle injection would
lead to greater oil recovery. First, the heavy oil saturated cores were subjected to a
secondary recovery process (waterflooding) until the water cut was 100%. Thereafter,
the nanoparticles aqueous dispersion was injected as a tertiary recovery to evaluate the
additional heavy oil recovery by the generation of a low viscosity o/w emulsion, and the
reduction of the interfacial tension between the injected and displaced fluids.
12
II. HEAVY OIL:
A VAST SOURCE OF ENERGY FOR THE FUTURE
2.1 Importance of heavy oil due to the natural depletion of conventional oil
Currently the rapid population growth demands for higher consumptions of
energy, principally of hydrocarbons. This situation associated with the natural depletion
of conventional oil production creates an unfavorable gap between oil produced and oil
required for the future. Because of this, the oil industry needs to move beyond the
development of conventional resources and produce hydrocarbons from unconventional
ones. Because heavy oil represents more than two-thirds of the world total resources
(Alboudwarej et al. 2006), this gap may be rapidly reduced by mastering heavy oil
recovery processes.
2.1.1 Definition of heavy oil
Heavy oil is an unconventional resource that is mainly characterized by its high
viscosity and high density when compared to conventional oil. High viscosity and high
density arise from the loss of most of the lighter hydrocarbons, leaving behind high
contents of asphaltenes, nitrogen, oxygen, sulphur and heavy metals (Attanasi and
Meyer 2010).
13
2.1.2 Classification of heavy oil
Literature does not present a formal classification of heavy oil; however, there
are some classifications accepted throughout the oil industry. The API classification is
probably the most commonly used. API gravity is defined as the measurement of the
relative density of oil when compared to water. API gravity is related to the specific
gravity of the oil through the Eq. 1.
P . . .
- . ....................................................................................... (1)
According to the API gravity definition, heavy oil has less than 22.3° API gravity
(Table 2.1). Despite providing a fair distinction among all types of crude oils, the API
gravity classification might not be optimal when used only to classify heavy oil. Its
application is limited since it does not consider the viscosity, a property that affects the
most heavy oil recovery and its transportation processes.
Table 2.1—Classification of crude oil according to API gravity.
Type of
crude oil
API gravity
°API
Light > 31.1
Medium 22.3 - 31.1
Heavy < 22.3
14
Another classification is given by the World Petroleum Congress where viscosity
is the principal discriminator, and it serves to categorize heavy oil into heavy and extra-
heavy or bitumen (Table 2.2).
Table 2.2—Classification of crude oil according to the World Petroleum Congress
API gravity
°API
Viscosity
cP
Heavy Oil 22 - 10 100 - 10,000
Extra heavy oil / Bitumen < 10 > 10,000
2.1.3 Degradation of crude oil to origin heavy oil
Different explanations have been proposed about the origins of heavy oil.
However, biodegradation seems to be the most widely accepted (Head et al. 2003). In
order to have a biodegradation process, light and medium oil must have been generated
and expelled from the source rock. Light and medium oil start a migration process away
from the source rock. Some hydrocarbons may travel updip for long distances, in the
order of the hundreds of miles, until they are trapped in shallow and cool reservoirs.
These reservoirs may have contact with meteoric water creating conditions for biological
activity (biodegradation). Biodegradation involves processes that oxidize the oil that
directly affects the composition and the physical properties of oil. Biodegradation might
decrease the density and increase the sulphur, acidity, viscosity and metal content (Head
et al. 2003). Degradation of crude oil may also be caused by physical processes such as
water washing or phase fractionation (Alboudwarej et al. 2006).
15
2.1.4 Estimation of global resources
Unlike conventional crude oil, heavy oil does not flow as easily, and it is
necessary to use special equipment or sophisticated recovery methods to extract it. This
increases both the production and refining costs and due to its high density and high
viscosity its economic value is also depreciated. Nonetheless, despite these undesirable
conditions, heavy oil has a great importance for the energy future because the estimated
volumes are vast. The IEA estimates the amount of heavy oil to be between 9 and 13
Tbbl of oil in place worldwide. This represents more than two-thirds of the conventional
resources (Alboudwarej et al. 2006) (Fig. 2.1). The Western Canada Basin in Alberta
with 2.5 Tbbl and the Orinoco Belt in Venezuela with 1.5 Tbbl are considered the largest
deposits in the world. In addition, significant accumulations of heavy oil have been
reported in other countries such as: the United States, Russia and Mexico.
Despite the fact that heavy oil resources are very vast, the current reserves are
still lower than conventional oil reserves. Conventional oil reserves amount tos 2.25
Tbbl while heavy oil ones are estimated around 1.88 Tbbl (IEA-WEO-2012). This
particular difference is because in-place hydrocarbon volumes can only be considered as
reserves if they are technically and economically recoverable under current conditions.
As the industry progresses in the improvement and development of new techniques for
heavy oil recovery, it will be able to reduce cost and increase production volumes.
Furthermore, if favorable oil prices are present, current heavy oil resources may be
reclassified as reserves.
16
Fig. 2.1—Importance of the heavy oil lies in the great volume in-place since it
represents 70% of the world total resources
2.2 Current status of the technologies of heavy oil recovery processes
Since the first discoveries of heavy oil reservoirs in the early twentieth century,
the main challenge has been focused on finding the optimal techniques to exploit them at
economically profitable rates that maximize their recovery factor. Throughout these
years, different recovery techniques have been developed ranging from mining pits to
more complex ones such as in-situ combustion. Normally, these types of deposits begin
their exploitation through primary production methods up to reaching their economic
limit. Thereafter, thermal methods are applied to reduce the oil viscosity and increase its
mobility allowing, additional recovery that is often far above the initial one. An
illustrative example is the Kern River oil field in Bakersfield, California. The Kern River
oil field is a heavy oil field producing since 1899 with a primary recovery of 5-10%.
17
This field was subjected to a steam-flooding process in 1965 and it is estimated to
achieve a final recovery of 80% (Nelder 2011).
Unfortunately, thermal methods cannot be applied to every reservoir; for
example, steam flooding is more efficient if it is applied to shallow depth and thick
reservoirs due to large heat losses (Falcone 2009). Other thermal methods, such as ISC,
may be applied beyond these limitations but to the best of our knowledge they have not
been sufficiently studied thus far.
Non-thermal EOR processes such as waterflooding, AS flooding, polymer
flooding and emulsion flooding may provide an excellent alternative to overcome depth
and thickness reservoirs issues. Moreover, these processes are less expensive than
thermal methods, and if favorable reservoir conditions exist, (e.g. adequate porosity,
permeability and a suitable oil viscosity) they may achieve high recovery factors.
Different technologies and EOR methods applicable to heavy oil have been
classified in to three categories Primary, Thermal and Non-Thermal (Fig. 2.2).
18
Fig. 2.2—Heavy oil recovery process may be divided in Primary, Non-Thermal and
Thermal methods
2.2.1 Primary recovery methods
These methods are mainly used in very shallow (<300 ft) or surface oil
accumulations such as the Canadian tar sands. These methods are also applied when
heavy oil can flow by itself as in Orinoco Oil Belt in Venezuela. Primary recovery
methods have low economic and technical risk and in most cases they are the first step
along the oil field development project (Shah et al. 2010). They are subdivided as: open
pit mining, high contact wells, and cold heavy oil production with sands (CHOPS).
Open pit mining
If the bitumen deposits are close to the surface, they can be recovered using
surface mining operations. This process consists in removing all the soil over the
bitumen and then excavating a pit in order to extract the hydrocarbons. Mining processes
19
may recover up to 90% of the total bitumen in place (Clark et al. 2007). However, they
are limited to maximum depths of 300 feet. Additionally, surface mining has a high
environmental impact and companies are subjected to land reclamations. This process
has only been used in Western Canada and Russia (Shah et al. 2010).
High contact wells: horizontal and multilateral
Horizontal and multilateral wells have been extensively used in the Orinoco
heavy oil belt in Venezuela due to the oil in-situ viscosity which is sufficiently low that
it can flow at reservoir temperatures. In addition, the horizontal sections and the multiple
branches are drilled to maximize contact area with the reservoir, allowing the well to
produce at economic rates (Fipke and Celli 2008). Progressive cavity pumping (PCP) or
electric submergible pumps (ESP) are almost always utilized along with horizontal and
multilateral wells to reduce the hydrostatic pressure in the wellbore and lifting the oil.
Cold heavy oil production with sand (CHOPS)
Dusseault (2002) defined CHOPS as the primary heavy oil recovery process that
consists of the intentionally production of sand along with the oil from thin and shallow
reservoirs. This method, combined with a progressive cavity pumping (PCP), has been
widely applied in Canada and Venezuela, increasing oil production rates significantly.
However, the application of CHOPS has to be carefully evaluated since the continuous
sand production causes the creation of high permeability networks or wormholes that are
unfavorable for future EOR operations. In addition, CHOPS has a recovery factor of 5 to
20
15% (Shokri and Babadagli 2012). Although, it is better than high contact wells, it is
outperformed by steam-flood methods. There are also concerns associated with
environmental problems due to sand disposal and possible subsidence effects that cannot
be ignored.
2.2.2 Thermal enhanced heavy oil recovery methods
Thermal EOR methods have as main objective to increase the mobility of the
viscous oil through the injection of heat into the reservoir, to raise the temperature, and
to reduce the oil viscosity. It may be carried out by injecting either steam or air to
combust the oil in-situ. The efficiency of these methods will depend on the heat losses in
the wellbore and the reservoir, and the heat transfer to the oil (Shell 2012). They are
subdivided as: Cyclic steam stimulation (CSS), steam flooding, steam assisted gravity
drainage (SAGD), and in-situ combustion (ISC).
Cyclic steam stimulation (CSS)
Cyclic steam stimulation, also known as “huff and puff”, is a single well
technique that involves three stages starting by the injection of high pressure steam into
the reservoir from 2 to 30 days. Then, the well is shut-in from 5 to 30 days to soak the
near-wellbore reservoir allowing the heat transference from the steam to the heavy oil.
Finally, the well is open to production, and the heated oil and the condensed water are
produced until the oil rate becomes uneconomic. The production stage may last up to
one year (Revana and Erdogan 2007). Once the cycle is completed, the process is
21
repeated as many times as the geological features and fluid properties allow producing
additional heavy oil. Recovery factors typically are in the range of 20% to 35% of OOIP
(Clark et al. 2007; Shah et al. 2010).
Steam flooding (Steam drive)
Steam flooding is a multi-well pattern technique that consists in the continuous
injection of steam into the reservoir to enhance recovery. The steam is used to heat up
the oil to reduce its viscosity and improve its mobility. In contrast to the CSS method,
this technique adds an energy drive mechanism by displacing the viscous oil towards the
producing wells. Despite the heat losses from the surface facilities to the reservoir and
the override effects caused by the density difference between the heavy oil and the
steam, the recovery factors may be high (Matthews 1983). There are many successful
projects, such as the Duri field in Indonesia and the Kern River field in California with
recoveries of 70% of the OOIP.
Steam assisted gravity drainage (SAGD)
Steam assisted gravity drainage is a steam based recovery method first conceived
by Dr. Roger M. Butler et al. (1981), which involves the drilling of two parallel
horizontal wells separated vertically from 16 to 23 ft. from one to another with an
horizontal displacement from 1,600 to 5,000 ft. long (Clark et al. 2007). The process
starts with the injection of steam into the reservoir through both wells to heat and
mobilize the oil between them, with the purpose of establishing communication (Butler
22
1994). Afterwards, steam is continuously injected by the injection well (upper). The
injected steam travels upwards into the reservoir forming a steam chamber and
transferring the steam’s heat by convection to the heavy oil; thus, it reduces the oil’s
viscosity. The lesser viscous oil flows downward (gravity influence) along with the
condensed water, where it is pumped out of the ground through the producer well
(lower). Because of the characteristics of the process, it is suited for extra heavy oil and
bitumen, as well as for high vertical permeability reservoirs. Expected recovery factors
are in the order of 50% to 70% (Alboudwarej et al. 2006).
In situ combustion (ISC)
In-situ combustion (ISC), or fire flood, is a method that consists of the
continuous injection of air into the reservoir to generate a combustion front where a
small portion of oil is burned. The generated heat reduces the viscosity of the unburned
heavy oil improving its mobility. Moreover, due to the high temperatures reached at the
combustion front, a fraction of heavy oil is in-situ upgraded by cracking (Alboudwarej et
al. 2006). As the combustion front moves forward to the production well, the forefront
heated oil (oil bank) is driven principally by combustion-gas drive as well as water and
steam drive (Sarathi 1999).
The process may be cataloged as forward in-situ combustion if the combustion
front starts close to the injector well and the combustion front moves in the same
direction of the injected air. If the combustion front starts close to the producer well and
23
it moves in the opposite direction of the injected air, it is known as reverse in-situ
combustion (Thomas 2008). In the reverse combustion, zones near the producer well are
initially heated to lower their viscosity to facilitate the flow of the oil that is beyond this
initially heated oil. However, this technique has only been tested successfully at
laboratory scale (Shah et al. 2010). Forward combustion is considered as “dry” if only
air is injected to maintain the combustion front, or “wet” if water is also injected along
the air to better displace the oil. According to Sarathi (1999), in wet combustion less
heavy oil is used as fuel. Moreover, it is more efficient since water provides better heat
transport and drive mechanism.
ISC has several advantages over steam based methods, for example it does not
require an energy source to generate the steam. In addition, it is not depth restrict and it
can be applied either to heavy or light oil reservoirs (Sarathi 1999). Additionally, there is
no heat loss from the surface to the reservoir and thermal efficiency within the reservoir
is high due to the in-situ nature of the process. Recovery factors greater than 50% have
been documented in the Suplacu de Barcau field, Romania and Bellevue field,
Louisiana, US (Turta et al. 2007). Nevertheless, this process has not been widely
implemented in the field because of the difficulty to control the combustion front,
coupled with the generation of flue toxic gases and gravity segregation of the
combustion front (Hincapie et al. 2011).
24
Toe to heel air injection (THAI) and catalytic upgrading process in situ (CAPRI)
Toe to heel air injection (THAI) is an ISC based technique. It was designed to
address common problems in conventional ISC, such as control of the combustion front,
gas override and air channeling. THAI is an innovative technology developed by Dr.
Malcolm Greaves at the University of Bath in England and patented by PETROBANK
in Canada. It combines a vertical air injector well, generally settled at the top of the
reservoir, and a horizontal producer well in a line-drive close to the bottom of the
reservoir (Greaves et al. 2001; Xia and Greaves 2006).
The process begins with a preheat stage where steam is injected in both the
vertical well and the horizontal well (“toe”) to establish communication between them.
Thereafter, air is injected in a controlled manner into the formation by the vertical well
to generate and propagate the combustion front throughout the horizontal well from the
“toe” to the “heel” (Thomas 2008). Temperatures over 600 °C are reached in the
combustion front reducing the oil viscosity and thermal oil cracking. Movable fluids
flow in a stretch zone called “the mobile oil zone” (MOZ), where they are driven by
gravity to the horizontal well (Xia et al. 2003).
Currently there are no commercial projects using THAI. However, laboratory
studies using Canadian heavy crude oil have exhibited recovery factors greater than
80%. Furthermore, heavy oil samples have experienced increments in the order of eight
API degrees with viscosity reductions from 100,000 cP to 50 cP during the ISC.
25
Taking advantage of the efficiency of THAI, a variation of it was developed in
order to upgrade the cracked oil even more. CAPRI acts as a subsurface refinery. It
consists in packing the horizontal section of the producer well with a catalyst to develop
a catalytic cracking process. Results obtained in laboratory show that is possible to
upgrade the produced oil from four up to six API degrees additional to the THAI
process. Moreover, a considerable reduction in the amount of heavy metals and
asphaltenes was observed. This process seems to be a good alternative as it enhances the
benefits of THAI, while reducing refining cost as well as environmental impacts
(Greaves and Xia 2001).
2.2.3 Non-thermal enhanced heavy oil recovery methods
According to the literature, non-thermal EOR methods have not been widely
used to recover heavy oil, mainly because these processes do not improve the fluid flow
characteristics of the oil as the thermal methods do. However, they should not be
dismissed because they may be used in deep and thin formations, and they eliminate the
energy requirement to generate heat, which leads to a reduction of capital costs and may
lead to the economic success of a project.
Waterflooding
Waterflooding is not a common recovery technique in heavy oil reservoirs. The
main problem associated to waterflooding in heavy oil is the significant difference
between the heavy oil viscosity and the water-injected viscosity. Whereas heavy oil
26
viscosity may be greater than 100 cP; injected water has a viscosity around 1 cP. This
difference causes an unfavorable mobility ratio, which results in oil bypassing due to the
viscous channeling through the reservoir.
Regardless of the adverse conditions to apply waterflooding in heavy oil
reservoirs, some projects have been implemented in Canada (Adams 1982; Miller 2005)
because waterflooding is a well understood technique and it is not as expensive as
thermal methods (Mai and Kantzas 2009). Estimated ultimate recovery from 20% up to
45% can be obtained if the heavy oils have an oil gravity greater than 15° API, viscosity
in the order of 100 cP and a well spacing less than 50 acres (Lu et al. 2010). However,
increasing the viscosity of the injected-water is recommended to obtain a better sweep
efficiency and improve recovery factors (Brook and Kantzas 1998).
Polymer flooding
Injecting water into heavy oil reservoirs may not deliver the recoveries because
of the adverse mobility ratio. Nonetheless, if the water-injected viscosity is increased by
adding a water-soluble polymer, (e.g. polyacrylamide or polysaccharide) mobility and
sweep efficiency may be improved resulting in higher recoveries (Asghari and Nakutnyy
2008). According to Needham and Doe (1987), polymer flooding may be more effective
than waterflooding due to the effects that the polymers have over fractional flow, as a
result of the improvement in the mobility ratio as well as the polymer flow through
unswept zones.
27
Polymer flooding has been more effective for medium oil than heavy oil
reservoirs. However, some heavy oil reservoirs have been treated with this process with
a relative level of success (Selby et al. 1989). Some limitations of this process are
increased injection pressure due to the polymer solution’s high viscosity as well as the
degradation of the polymers in the rock with concomitant loss of viscosity/mobility of
the drive fluid. Laboratory tests show high recoveries, up to 70% with a polymer
concentration of 10,000 ppm in 1% of brine solution and a heavy oil of 1,000 cP
(Asghari and Nakutnyy 2008).
Alkali-surfactant (AS) and alkali-surfactant-polymer (ASP) flooding
Alkali-surfactant flooding is a promising method to increase heavy oil recovery.
It is based on the reduction of the interfacial tension (IFT) between heavy oil and
injected water by injecting an alkali/surfactant solution into the reservoir. The reduction
of the IFT allows the generation of emulsions responsible for the heavy oil recovery.
Laboratory studies have shown the in-situ creation of w/o and o/w emulsions.
Since, w/o emulsions are more viscous than the original oil, the displacement front
created during the flooding improves the sweep efficiency (Ali et al. 1979). Conversely,
other researchers suggest that the external water phase emulsions can emulsify the heavy
oil and produce it as an o/w emulsion (Bryan and Kantzas 2009). A third view involves
the in-situ generation and the interaction of w/o and o/w emulsions into the reservoir. In
this view, w/o emulsions provide the drive mechanism while the o/w emulsions plug off
28
the high permeability channels reducing the viscous fingering and improving the sweep
efficiency (Bryan and Kantzas 2008; Mai et al. 2009).
The ASP process adds a polymer to increase the viscosity of the AS solution to
reduce the water channels and give a better mobility control (Arihara et al. 1999). Pilot
tests have been performed in Daquing oil field in China. This test obtained recovery
factors up to 20% of the OOIP (Shutang and Qiang 2010). This process improves the
results given by alkali/surfactants; however, the high capital costs limit its commercial
use.
Vapor extraction (VAPEX)
VAPEX takes the concept of SAGD but replaces the steam injection with
vaporized hydrocarbon solvents (e.g. ethane, propane and butane). VAPEX reduces the
heavy oil viscosity by diluting it with vaporized hydrocarbon solvents in order to
increase mobility of the oil (Das 1998). As the solvent vapor is injected, it creates a
vapor chamber over the horizontal injector and it is diffused and absorbed by the heavy
oil. Consequently the heavy oil viscosity is reduced and upgraded by deasphalting
(Upreti et al. 2007). The mobile oil drains to the horizontal producer by gravity. The
process continues until it reaches the economic limit. Compared to SAGD, the VAPEX
process is energy and economically efficient because it does not require a great amount
of energy. This reduces a great deal of the capital and operational costs (Pourabdollah
29
and Mokhtari 2013). The major constraint of this process is the slow diffusion of the
solvent resulting in low production initial rates (Upreti et al. 2007).
Carbon Dioxide (CO2) injection
CO2 injection has been widely used in the U.S. as an EOR process in medium
and light oil reservoirs. Due to its high miscibility with those types of oil at supercrital
and liquid phase, the oil swells and its viscosity is reduced, improving the final recovery.
Moreover, this process can also be seen as a potential alternative to store CO2 produced
by anthropogenic sources, such as the coal based power plants (Sohrabi et al. 2007).
Unlike medium and light oil, miscible CO2 injection in heavy oils becomes more
difficult since it requires high injection pressures to reach the minimum miscibility
pressure. However, immiscible injection has been seen as a solution to this problem.
CO2 gas injection in heavy oil has been studied by several authors. They have stated that
CO2 in gas phase is highly soluble in heavy oil reducing considerably the viscosity
(Emadi et al. 2011).
Despite heavy oil’s viscosity reduction, CO2 injection in a gas phase by itself
cannot produce commercial volumes of heavy oil. The great contrast in viscosity
between fluids causes an adverse mobility ratio and therefore a poor conformance
(Zhang et al. 2006). This process can be improved by alternating injection of CO2 and a
higher viscous fluid. Technologies such as water alternating gas (WAG) (Zheng and
30
Yang 2012), chemical alternating gas (CAG) (Zhang, Y. et al. 2010) and foam injection
(Emadi et al. 2011) have shown favorable results, making these techniques potential
alternatives for heavy oil recovery.
31
III. EXPERIMENTAL DESCRIPTION
3.1 Experimental materials
3.1.1 Modified silica hydrophilic nanoparticles
Modified silica hydrophilic nanoparticles of 5 nm and 20 nm mean diameter
were received from 3M Company as an aqueous suspension of 20 wt%. These
nanoparticles were treated with alkyl ether to give them a hydrophilic nature. Because of
this characteristic, the hydrophilic nanoparticles are able to stabilize o/w emulsions
without the addition of surfactants.
3.1.2 Heavy crude oil
Two different heavy oils samples were used in this study. The first sample is
from a Louisiana Oil field and the second is from the Kern River oil field in California.
Table 3.1 shows the available properties of each sample.
Table 3.1—Properties of the two heavy oils used in this research
Property
Louisiana
Heavy Oil
(LA HO)
Kern River
Heavy Oil
(KR HO)
Density [g/cm3] 0.941 0.944
API Gravity [°] 18.70 18.31
Viscosity [cP] @ 72°F 600 3,500
32
3.1.3 Sandstone cores
Two different permeability sandstone cores were utilized in the flooding
experiments. The effect of permeability on the emulsion generation and the final
recovery was evaluated in these cores. The first type of core was Buff-Berea sandstone
with a permeability that ranged from 120 to 350 mD and a porosity of 22%. The second
core was Bentheimer sandstone with a permeability of 2,000 to 2,540 mD and 24% of
porosity.
3.1.4 Brine solution
A 5,000 ppm brine solution was prepared using deionized water and 5.0 wt% of
sodium chloride (NaCl). This solution was made with the aim of measuring the core’s
porosity and permeability; in addition, it was used as displacement fluid in the
waterflooding experiments and as a solvent of the nanoparticle dispersions.
3.2 Emulsion generation and properties bench test
3.2.1 Emulsification
An emulsion is generally defined as a mixture of two immiscible substances (e.g.
oil and water) where one substance is dispersing in the other. The application of shear
through mechanical means is usually necessary to generate an emulsion. In this study
bulk emulsions were prepared to determine optimum loading of nanoparticles to
33
minimize shear required for emulsion formation, using the following equipment and
methodology.
Stirrer
A CAFRAMO model RZR50 stirrer was used to generate the emulsions. See Fig.
3.1. It operates by rotating a blade propeller at high speeds up to 2,000 rpm. The
emulsification procedure consisted in a 150 ml beaker filled with 25 ml of nanoparticle
dispersion and 25 ml of heavy oil with the stirrer’s propeller fully immersed. The
propeller was positioned off the center of the beaker to reduce the vortex formed by the
centrifugal forces which appeared to inhibit emulsion generation. A shearing time of 10
minutes was established to generate the emulsions.
Fig. 3.1—CAFRAMO RZR50 stirrer
34
3.2.2 Properties measurement
Essential properties of generated emulsions such as viscosity and interfacial
tension were measured. The viscosity of the emulsion is the most important property in
this process since the o/w emulsions trap the viscous oil transporting it out of the rock as
a lower viscosity fluid. The emulsion samples were prepared as is stated in section 3.2.1.
Evaluating the reduction of the interfacial tension between heavy oil and nanoparticle
dispersion is another important aspect of the emulsions generation that may impact
ultimate oil recovery.
Viscosity measurements
A Brookfield DV-III Ultra Programmable Rheometer (Fig. 3.2) was used to
measure the fluids’ viscosity. Additionally, an enhanced UL adapter was attached to the
rheometer to probe small amounts of fluid. The UL adapter consists of a sample chamber
and a spindle. The rheometer calculates the shear stress at a given shear rate; these
values are plotted in a Cartesian “xy” graph where the resulting slope is the viscosity.
Interfacial tension measurements
Interfacial tension (IFT) was measured by the pendant drop analysis using an
Optical Contact Angle Measurement System OCA 15Pro (Fig. 3.3). The OCA 15Pro
software automatically calculates the IFT fitting the drop shape image to the Laplace-
Young equation.
35
Fig. 3.2—Brookfield DV-III Ultra Programmable Rheometer
Fig. 3.3—Interfacial tension System OCA 15Pro
36
3.3 Core flooding system
Multiple core flooding experiments were carried out to evaluate the in-situ
emulsion generation as well as prove that the nanoparticles can be used as an EOR
method to recover heavy oil. The core flooding system is integrated with the following
elements: a Teledyne ISCO syringe pump and a heavy oil accumulator, a core holder and
a hydraulic pump to generate confinement pressure. These elements were interconnected
with / 8” stainless steel tubing. See Fig. 3.4.
Fig. 3.4—Actual and schematic core flooding system
The Teledyne ISCO syringe pump along with the accumulator were used as the
injection system. When brine or nanoparticle dispersion was injected directly into the
core, the accumulator was isolated from the system through the valves. To saturate the
core with heavy oil, valves were switched so that the syringe pump was connected to the
accumulator inlet; the accumulator outlet fed the inlet to the core holder.
37
The core holder is fabricated in aluminum and it is able to support pressures up to
3,000 psi. It is designed to hold cores with a maximum length of 1 ft. and 1 in. of
diameter. It is comprised of three sections: an inlet cap with an adjustable plunger, this
plunger should be adjusted until it is in contact with the core, a main cell body where the
core is placed, and an outlet cap with a fixed plunger. In this research, cores of 6 in.
length were positioned in the core holder abutted by a series of three perforated
aluminum plugs to span the core holder length. The cores and the plugs were placed in a
rubber sleeve and secured at the ends by plungers. A hydraulic pump was used to fill the
annulus between the rubber sleeve and the main cell body with hydraulic oil;
additionally, it was also used to apply the overburden pressure over the rubber sleeve.
3.4 Experimental procedures
To achieve the objectives of this research, the experiments were divided in two
stages: first the mechanical emulsion generation, then the core flooding tests. Both
stages were performed at ambient temperature. Procedures describing how these stages
were conducted are detailed below.
3.4.1 Mechanical emulsion generation
To generate the emulsions, 25 ml of crude oil were placed in a 250 ml beaker in
which the stirrer was run at 2,000 rpm. Then 25 ml of the nanoparticle dispersion was
poured in small quantities. After 10 minutes in the stirrer, the emulsions were collected
and stored in sample vials to be analyzed afterwards.
38
3.4.2 Core flood experiment procedures
In order to prove the enhanced recovery potential by injecting hydrophilic
nanoparticle dispersions, a series of core flooding tests was performed with two different
types of heavy oil and two kinds of rocks with different permeabilities. The test
procedure consists of two parts: (1) characterization of rock porosity and permeability;
and (2) core flood measurements of secondary and tertiary recovery.
Porosity
Core porosity was calculated by measuring the bulk volume and the pore volume
of the core. This was done following the steps shown below.
1. Bulk volume ( ) was calculated assuming that the cores have cylindrical shapes
and using Eq. 2. Given that core plugs are not uniform, averages of diameters and
length were calculated. For the diameter, measures of ten different points were
taken, and for the length five measures were taken both using an electronic
Vernier Caliper.
................................................................................................ (2)
2. Pore volume ( ) was calculated with the liquid saturation method represented in
Eq. 3. Dry weight ( ) of the core plugs was measured after placing them in an
oven for 24 hours to remove any liquids present in the pores. In order to obtain the
39
wet weight ( ) of the core plugs, they were saturated with brine and weighed.
To guarantee 100% saturation, core plugs were placed in a desiccator full of brine
and connected to a vacuum for 24 hours. Brine density was measured using a
densimeter.
......................................................................................... (3)
3. Finally, porosity was calculated as the bulk volume and pore volume ratio, as
depicted in Eq. 4.
................................................................................ (4)
Brine permeability
Permeability to various fluids (k) was calculated by applying Darcy’s law, as
described in Eq. 5.
................................................................................................. (5)
The permeability calculation procedure consists of the following steps:
40
1. Place the brine saturated core plug in the core holder, and apply a confining
pressure of 1,800 psi with the manual pump.
2. Set up the syringe pump to the core holder and start the brine injection at a specific
rate ( ) until the steady state is reached. At this moment the pressure drop ( )
value is registered. This step was done for several different injection rates.
3. Plot vs values and draw a trend line (Fig. 3.5).
4. The value of the trend line slope represents the value of the core plug permeability
to brine.
Irreducible water saturation (Swirr) and initial oil saturation (Soi)
Before starting the secondary and tertiary recovery test; the core plugs must be
prepared to the initial conditions of oil and irreducible water saturations. To obtain these
values the next procedure was followed:
1. Once the core plug is 100% saturated with brine and installed in the core holder;
the syringe pump is connected to the oil accumulator inlet, which is also
connected to the core holder via its outlet.
2. Crude oil injection starts at a rate equivalent of a Darcy velocity of 1 ft/d.
3. Keep the injection rate until the water production ends.
4. Initial oil and irreducible water saturations are calculated with material balance.
41
Fig. 3.5—Core permeability to brine was measure at different injection rates and
its corresponding pressure drop.
Waterflooding (Secondary recovery)
Once the core plug has been saturated with heavy oil until the irreducible water
saturation is reached the secondary recovery process is started. A brine of 5,000 ppm
NaCl is used as the displacement fluid. The waterflooding process was completed
following the next steps:
1. Place the brine directly in the syringe pump and connect it to the core holder.
2. Start the brine injection at a rate equivalent of a Darcy velocity of 1 ft/d.
3. The brine is injected until the water cut is 100%.
4. The heavy oil recovery is calculated using a material balance.
y= 271x mD
R² = 1
42
Nanoparticle dispersion flooding (EOR-Tertiary recovery)
In order to achieve the research objectives; an EOR process was performed after
the waterflooding process. The EOR consists of the injection of a hydrophilic
nanoparticles dispersion, which function is to emulsify the residual heavy oil and to
facilitate its extraction from the core plug. The next steps describe the procedure
followed in this stage:
1. Place the nanoparticle dispersion in the syringe pump and connect it to the core
holder.
2. Start the nanoparticle injection with a low injection rate and increment it until a
rate where emulsions generation is reached. Injection rates equivalent of a Darcy
velocity from 1 to 500 ft/d were set.
3. The nanoparticle dispersion is injected until the heavy oil or emulsion production is
zero.
4. Collect the produced fluids in a vial that has a tightly-closeable lid, and place them
in an oven for 24 hours at a temperature of 85° C to break the emulsion and
measure the oil volume.
43
IV. ANALYSIS OF RESULTS
Emulsion generation experiments were carried out using a stirrer as the shear rate
provider. Six different nanoparticle dispersions were tested to investigate the effect of
the nanoparticle and NaCl concentrations in the generation of emulsions. Emulsions
were created by mixing the nanoparticle dispersions with two types of heavy oil.
Subsequently, a bench test analysis of each emulsion was performed in order to select
the most appropriate dispersion to be used as the injection fluid in the core flood
experiments. Six core flood experiments were performed using the optimal dispersion,
previously identified. The optimal dispersion yielded an incremental recovery, providing
proof-of-concept that nanoparticle dispersion may be used as an injection fluid in an
EOR process.
4.1 Emulsion generation
Six nanoparticle dispersions were created using nanoparticles concentrations of
0.5, 2.0, and 5.0 wt%. Three of them were made using deionized water, and the other
three were made with a 0.5 wt% NaCl brine, with the aim to investigate the effects of
nanoparticles and NaCl in the emulsification process.
The six dispersions were mixed with two types of heavy oil, described in the
previous chapter, using a stirrer set at 2,000 RPM for ten minutes. Twelve samples were
obtained as a result of this process. Additionally, four more tests were made using
44
deionized water and brine without adding nanoparticles in order to establish the base
cases of this experimentation process (Table 4.1). Base line samples containing no
nanoparticles and no or 0.5% NaCl solution were also evaluated.
Table 4.1— Sixteen emulsion generation experiments were conducted to identify
the optimum nanoparticle concentration to generate the largest amount of
emulsions
Crude Oil
(cP)
Nanoparticle
concentration
(wt%)
Salinity
(wt% NaCL)
1 600 0.0 0.0 2 600 0.5 0.0 3 600 2.0 0.0 4 600 5.0 0.0 5 600 0.0 0.5 6 600 0.5 0.5 7 600 2.0 0.5 8 600 5.0 0.5 9 3,500 0.0 0.0 10 3,500 0.5 0.0 11 3,500 2.0 0.0 12 3,500 5.0 0.0 13 3,500 0.0 0.5 14 3,500 0.5 0.5 15 3,500 2.0 0.5 16 3,500 5.0 0.5
45
The experiments using deionized water and the Louisiana heavy oil showed a
rapid phase separation due to density differences; emulsion generation was not observed.
However, when the nanoparticles were added and their concentration was increased, an
emulsified heavy oil fraction formed and increased as a fraction of total fluid with
increasing nanoparticle loading. Similar results were observed with the nanoparticle-
plus-brine dispersions, but with a larger emulsified heavy oil fraction (Fig. 4.1).
Kern River heavy oil base cases also showed phase separation without emulsion
generation. Emulsions samples from the nanoparticle dispersion and this type of heavy
oil have similar tendencies to those obtained with the first heavy oil. However, samples
made with brine greatly increased the emulsified heavy oil fraction (Fig. 4.2).
Fig. 4.1—Samples of emulsions generated using deionized water and brine with
Louisiana heavy oil
46
Fig. 4.2—Sample of emulsions generated using deionized water and brine with
Kern River heavy oil
4.1.1 Nanoparticle dispersions and heavy oil interfacial tension
Interfacial tension (IFT) is an important property for the emulsion generation. If
the interfacial tension is significantly reduced, the emulsification process may be
achieved easily. Experiments were conducted to prove the hypothesis that the
hydrophilic nature of the nanoparticles, due to their alkyl ether surface, may cause a
reduction in the ITF since the alkyl ether works as a surface-active agent.
The pendant drop method was used to measure IFT. The pendant drop method
showed that the IFT between the heavy oil and the nanoparticle dispersion is reduced as
the nanoparticle concentration increases. In addition, if the salinity also increases, the
47
IFT reduction is even larger (Fig. 4.3). IFT was reduced when the nanoparticle and brine
concentration were incremented.
Fig. 4.3—IFT measurements show the reduction of IFT when the nanoparticle and
salt concentration is increased
The different concentration of nanoparticles changed the shape of the oil drop from a
rounded-like drop to and rope-like drop (Fig. 4.4 and Fig. 4.5). Rounded-like shapes were
observed for lower concentrations due to an increment in the IFT. The opposite was
observed for rope-like shapes in which higher concentrations yielded a reduction in the IFT.
48
The IFT for sample 16 (5.0 wt% of nanoparticle concentration, 0.5 wt% of NaCl
and Kern River heavy oil) was too low to be measured. The reduction in IFT did not
allow an oil drop to be formed, but a “rope-like oil” was formed instead, indicating a
very low IFT. n fact it is clear that the Kern River oil’s FT is much more sensitive to
increasing nanoparticle and brine loading than is the Louisiana oil at all additive
loadings. In contrast, in deionized water the Kern River oil exhibited the higher IFT.
Fig. 4.4—IFT between Louisiana heavy oil and the nanoparticle dispersions. A
reduction in the IFT is observed according to the nanoparticle and NaCl
concentration are increased
49
Fig. 4.5—IFT between Kern River heavy oil and the nanoparticle dispersions. IFT
was not measured in the final experiment since the oil drop was not formed
4.1.2 Microscopic emulsion imaging
Microscopic images of the emulsions were taken to prove the generation of
emulsions as well as to see the effects of nanoparticles and brine concentration (Fig. 4.6
and Fig. 4.7).
A greater number of emulsions can be seen in samples 12 and 16 as a result of
the high nanoparticle concentration. Hence, the maximum emulsion generation was
observed with the higher nanoparticle concentration. The emulsions made with the
Louisiana heavy oil did not form as many emulsion as the emulsions made with the Kern
River heavy oil.
50
Fig. 4.6—Microscopic images of emulsions made with Louisiana heavy oil and the
nanoparticle dispersions
Fig. 4.7—Microscopic images of emulsions made with Kern River heavy oil and the
nanoparticle dispersions
51
4.1.3 Emulsion rheology measurements
The rheology study of the emulsions plays an important role in this research.
Since the o/w emulsions are characterized by their water external phase, emulsion
viscosity should be close to that of water, at least for modest oil loadings in the
emulsion. Measurements can be done to estimate the emulsion’s viscosity and relate it to
water’s to assess the emulsions.
Results obtained from the rheometer show that, the emulsions created as it is
described in section 3.2.1, have a shear-thickening (dilatant) behavior. This indicates that the
apparent viscosity increases as the shear rate also increases. This behavior follows a power
law model (Eq. 6).
............................................................................................. (4)
Where is the apparent viscosity of the fluid [SI unit Pa·s], is the flow
consistency index [SI unit Pa·sn], is the shear rate [SI unit s-1] and is the
dimensionless power law index.
All of the emulsion samples behaved as a low viscosity fluid. Emulsions made with
5.0 wt% of nanoparticles showed higher apparent viscosities at higher shear rates. In
addition the more viscous heavy oil generated the more viscous emulsions at the same
nanoparticle concentration. The effect of the NaCl in the apparent viscosity was more
52
evident in the samples made with the Kern River oil and a concentration of 5.0 wt% of
nanoparticles.
In Fig. 4.8 illustrates the effect of the nanoparticle concentration in the emulsion
viscosity. As more nanoparticles were added, the apparent viscosity tends to be higher.
On the other hand, Fig. 4.9 shows the effect of the NaCl in the emulsion
viscosities. Small increments in apparent viscosity were noticed in samples 7, 8, 15
compared with the samples without NaCl; however, sample 16 showed a larger
increment.
Fig. 4.8—Increments in nanoparticle concentration leads to an increment in the
apparent viscosity; moreover, this viscosity is very low compared with the original
heavy oil
53
Fig. 4.9—Increments in apparent viscosity as a result of add NaCl were more
noticeable in sample 16
54
4.1.4 Selection of the nanoparticle dispersion for further EOR experiments
The pendant drop experiments showed an IFT reduction as the nanoparticle
concentration increased. When NaCl was added this reduction was further increased.
Microscope images confirmed that a larger number of emulsions were generated when
the IFT was lower. Finally, rheology experiments showed that the nanoparticles
emulsions behave as low viscous fluid.
The dispersion made with 5.0 wt% of nanoparticle and 0.5 wt% of NaCl
concentrations was selected to perform the core flooding experiments due to its emulsion
generation capacity and ITF reduction. Additionally, the dispersion made with a 2.0 wt%
of nanoparticle and 0.5 wt% of NaCl was also tested because if the results are favorable
in terms of recoveries, the reduction in nanoparticle concentration would improve the
project economics.
55
4.2 Core flooding experiments
Six core flooding (CF) experiments were carried out to validate the ability of the
hydrophilic nanoparticles to generate o/w emulsions in the porous media as well as to
assess the final recovery after the nanoparticle dispersion flooding.
Two different types of sandstone were used in these experiments to explore the
effect of permeability in emulsion generation. The cores were cylindrical-shaped, 6
inches long with a radius of 1 inch (Fig. 4.10). The upper core is a Buff Berea sandstone
and the lower is a Bentheimer sandstone. The Buff Berea sandstone core has a
permeability that ranges from 120 to 350 mD while the permeability of the Bentheimer
sandstone core ranges from 2,000 to 2,400 mD.
Fig. 4.10—Buff Berea and Bentheimer sandstone cores used in the flooding
experiments
56
To set up the original saturation conditions, and also to calculate the rock-fluid
properties for the previously described cores, the procedure outlined in section 3.4.1 was
followed; results are depicted in the Table 4.2.
Table 4.2—Rock-fluid properties and original saturation conditions
Core flooding (CF) 1 2 3 4 5 6
Sandstone type BB (I) BT (I) BB (II) BT (II) BB (III) BT (III)
Pore Volume [cm3] 16.5 18.0 16.2 18.1 15.9 17.3
Porosity [%] 22.0 24.2 21.6 24.1 21.1 24.1
Permeability [mD] 137.0 2,340.0 152.0 2,015.0 271.0 2,027.2
Swirr [%] 25.1 14.7 23.8 13.9 22.0 14.0
Soi [%] 74.9 85.3 76.2 86.1 78.0 86.0
OOIP [cm3] 12.4 15.4 12.4 15.6 12.4 14.9
Core flooding experiments were divided in two stages. The first stage was a
waterflooding process, where brine (see chapter 3) was used as the injected fluid. The
second stage was the nanoparticle dispersion flooding; in this stage the nanoparticle
dispersions were injected at different injection rates of 1 ft/d, 10 ft/d, 100 ft/d, 200 ft/d
and 300 ft/d. In this way the necessary shear rate to generate the emulsions was
achieved. CF1, CF2, CF3 and CF4 were completed using a nanoparticle dispersion made
with 5.0 wt% of nanoparticles and 0.5 wt% NaCl brine (ND1). CF5 and CF6 were
accomplished with the nanoparticle dispersion made with 2.0 wt% of nanoparticles and
0.5 wt% NaCl brine (ND2).
57
4.2.1 Effects of the ND1 on the in-situ emulsion generation and oil recovery
In this section, the results obtained from the CF1, are presented and discussed
next. The ability of ND1 to produce additional heavy oil, after waterflooding, as o/w
emulsions have been tested at different scenarios of rock permeability and oil viscosity.
Results are presented below.
Core flooding 1: Louisiana heavy oil – Buff Berea sandstone (I)
The Core flooding 1 was carried out with a Buff Berea sandstone core saturated
with Louisiana heavy oil. Absolute permeability of this core was 137 mD with a 22%
porosity and an OOIP of 12.4 cm3. During the waterflooding stage, brine was injected at
1 ft/d; this lasted until oil production was no longer observed; a total of 3 PV of brine
was injected, resulting in a recovery of 32 % of the OOIP.
ND1 was injected after waterflooding at different injection rates, resulted in an
additional 32.3% oil recovery; the total oil recovery at the end of the experiment was
64.3%. Table 4.3 shows a summary of the obtained results.
58
Table 4.3—Core flooding 1 experiment results
Core flooding 1
Heavy Oil type LA HO
Nanoparticles [wt%] 5.0
Sandstone type BB (I)
Oil Recovered WF [ cm3] 3.9
RF after WF [%] 32.0
Oil Recovered NF [ cm3] 4.0
RF after NF [%] 32.3
Final RF [%] 64.3
Fig. 4.11 shows the oil recovery and pressure profiles from CF1. It is observed
that after 2.5 PV of injected brine the oil recovery remains constant, meaning there is no
additional oil production. Furthermore, a maximum pressure of 105 psi is first observed
followed by a 35 psi pressure drop at the end of waterflooding.
59
Fig. 4.11—Oil recovery and pressure profiles along the CF1
Thereafter, the ND1 was injected at 1, 10, 100, 200 and 300 ft/d for 0.25, 1.5, 4.0
and 6.0 PV respectively. Note that at lower injection rates no additional oil was
produced. In addition at 100 ft/d, 200 ft/d and 300 ft/d crude oil was produced both as an
emulsion and as a single phase. A substantial increment in pressure was observed at high
rates because of the low rock permeability. Despite the significant oil recovered, a great
volume of dispersion had to be injected due to the high injection rates needed to generate
the emulsions. Table 4.4 shows the results of the dispersion flooding. Effluents
collected in this experiment are shown in Fig. 4.12. It was observed that ND1 was able
to produce emulsions and crude oil.
60
Table 4.4—Recovery factors and injection pressures at each injection rate, CF1
Injection
Rate
(ft/d)
Recovery
Factor
(%)
PV
Injected
Pressure
(psi)
o/w
Emulsions
1 0.0 0.25 37.0 no 10 0.0 1.5 91.0 no 100 4.0 4.0 210.0 yes 200 8.1 4.0 2,100.0 yes 300 20.2 6.0 2,700.0 yes
32.3 15.8
Fig. 4.12—Effluents collected from the CF1
61
Core flooding 2: Louisiana heavy oil – Bentheimer sandstone (I)
The Core flooding 2 was performed with a Bentheimer sandstone core saturated
with Louisiana heavy oil. Absolute permeability of this core was 2,340 mD, porosity of
24.2% and an OOIP of 15.4 cm3. After injecting 2.0 PV of brine the water cut was 100%
and the recovery factor was 24.1 % of the OOIP.
During the nanoparticle flooding, there was not emulsion generation; however,
additional single phase oil was produced. The additional recovered oil was 39% of the
OOIP with a final recovery of 63% of the OOIP. Table 4.5 summarizes the results of the
CF2.
Table 4.5—Core flooding 2 experiment results
Core flooding 2
Heavy Oil type LA HO
Nanoparticles [wt%] 5.0
Sandstone type BT (I)
Oil Recovered WF [ cm3] 3.7
RF after WF [%] 24.1
Oil Recovered NF [ cm3] 6.0
RF after NF [%] 39.0
Final RF [%] 63.1
62
The results in Fig. 4.13 show an important oil recovery at high injection rates
without pronounced increments in pressure; however, due to the high permeability of
this core enough shear rate values were not achieved to generate emulsions. Oil
production in this experiment is related mainly to an IFT reduction.
Fig. 4.13—Oil recovery and pressure profiles along the CF2
63
Table 4.6 presents the recovery factors of the nanoparticle flooding at each
injection rate as well as its corresponding injection pressure. Injection rates of 1 ft/d and
10 ft/d did not produce additional oil. At 300 ft/d, 400 ft/d and 500 ft/d, crude oil was
produced as a single phase and there was not emulsion generation (Fig. 4.14).
Table 4.6—Recovery factors and injection pressures at each injection rate, CF2
Injection
Rate
(ft/d)
Recovery
Factor
(%)
PV
Injected
Pressure
(psi)
o/w
Emulsions
1 0.0 0.25 1.0 no 10 0.0 0.75 3.2 no 300 7.8 4.2 9.5 no 400 14.3 5.5 10.4 no 500 16.9 7.0 12.6 no
39.0 17.7
Fig. 4.14—Effluents collected from the CF2.
64
Core flooding 3: Kern River heavy oil – Buff Berea sandstone (II)
The Core flooding 3 was carried out with a Buff Berea sandstone core saturated
with Kern River heavy oil. The brine permeability of this core was 152 mD with a
porosity of 21.6% and an OOIP of 12.4 cm3. This experiment was the most difficult to
complete since high permeability and high viscosity complicate the oil saturation and
flooding processes. Throughout the waterflooding stage, the injection pressure was in the
range of 2,000 psi range, reaching peaks of 3,500 psi, which is over the pressure limit of
the core holder. Waterflooding was performed at an injection rate of 1 ft/d; Oil
production was not observed after 2.5 PV of injected brine. The recovery factor after
waterflooding was 27.3% of the OOIP. Table 4.7 summarizes the results of this
experiment.
Table 4.7—Core flooding 3 experiment results
Core flooding 3
Heavy Oil type KR HO
Nanoparticles [wt%] 5.0
NaCl [wt%] 0.5
Sandstone type BB (II)
Oil Recovered WF [ cm3] 3.4
RF after WF [%] 27.3
Oil Recovered NF [ cm3] 2.5 RF after NF [%] 20.2
Final RF [%] 47.5
65
Fig. 4.15 clearly shows how the injection pressure was above 2,000 psi at all
times. The arrangement of high permeability and high viscosity led to the generation of
emulsion at low injection rates. However, it was not possible to continue the experiment
for longer periods of time due to the pressure limitations of the core holder.
Fig. 4.15—Oil recovery and pressure profiles along the CF3
66
In this experiment, the nanoparticle dispersion was injected at 1 ft/d, 5 ft/d and
10 ft/d. Despite the low rates, there was oil production as emulsion and as a single phase
due to a combination of high shear rates and an IFT reduction between heavy oil and the
injected fluid (Fig. 4.16). Table 4.8 details the results from this core flooding
experiment.
Table 4.8—Recovery factors and injection pressures at each injection rate, CF3
Injection
Rate
(ft/d)
Recovery
Factor
(%)
PV
Injected
Pressure
(psi)
o/w
Emulsions
1 2.6 0.27 2,065 yes 5 4.2 0.59 2,900 yes 10 9.1 0.9 3,500 yes 15.8 1.8
Fig. 4.16—Effluents collected from the CF3
67
Core flooding 4: Kern River heavy oil – Bentheimer (II)
Core flooding 4 was carried out with a Bentheimer sandstone core saturated with
Kern River heavy oil. Brine permeability of this core was 2,015 mD with a 24.1%
porosity and an OOIP of 15.6 cm3. The waterflooding stage was completed with a brine
injection rate of 1 ft/d obtaining a recovery factor of 23% of the OOIP. Table 4.9 gives a
summary of the results from CF4.
Table 4.9—Core flooding 4 experiment results
Core flooding 4
Heavy Oil type KR HO
Nanoparticles [wt%] 5.0
NaCl [wt%] 0.5
Sandstone type BT (II)
Oil Recovered WF [ cm3] 4
RF after WF [%] 23
Oil Recovered NF [ cm3] 10
RF after NF [%] 64.1
Final RF [%] 87.4
68
The recovery factor and pressure profile of this experiment are shown in Fig.
4.17. High permeability of this sandstone core did not present a flow restriction; hence,
there apparently was not enough shear to generate emulsion—at least there were no
emulsions observed at the core outlet sample collector. Despite this fact, there was
additional oil production in single phase; the recovery factor obtained by the
nanoparticle dispersion flooding was the highest of the six experiments. The IFT
reduction, observed in the pendant drop experiments, caused the oil production in this
experiment.
Fig. 4.17—Oil recovery and pressures profile along the CF4
69
Results shown in the Table 4.10 indicate that at higher injection rates, the
amount of produced oil was also higher. The injection pressure did not increase
significantly when the injection rates were incremented. In this experiment not only the
oil production was the largest, but the injected fluid volume was the smallest; no
emulsions were observed (Fig. 4.18).
Table 4.10— Recovery factors and injection pressures at each injection rate, CF4
Injection
Rate
(ft/d)
Recovery
Factor
(%)
PV
Injected
Pressure
(psi)
o/w
Emulsions
1 0.0 2.2 1.0 no 50 3.2 2.2 5.3 no 100 9.6 2.2 7.5 no 200 9.6 2.2 9.5 no 300 12.8 2.2 10.5 no
400 28.8 2.2 11.0 no
35.2 11.0
Fig. 4.18—Effluents collected from the CF2
70
Effect of permeability on the final oil recovery
The effect of permeability in any EOR process is fundamental for the success of
the project. In addition, permeability of the rock is essential to select the optimum
recovery process. The effect of permeability was investigated in this research to
determine how it affects the heavy oil recovery during the nanoparticle dispersion
flooding.
Effect of permeability on the lower viscosity heavy oil
Fig.4.19 shows a comparison of the recovery factor obtained from BB (I) and BT
(I) sandstone cores. Brine permeabilities were 137 mD and 2,340 mD respectively.
These cores were saturated with Louisiana heavy oil of 600 cP. Recovery factor after
waterflooding was larger in the BB (I) sandstone because the lower permeability reduced
the water channeling, improving the sweep efficiency. During the nanoparticle flooding,
emulsions were not generated in the higher permeability core due to the lack of shear
developed in the rock during flooding; it is believed that oil production from this rock is
only due to a reduction in the ITF. However, final recoveries were similar in both
experiments. The effect of permeability is also observed in Fig. 4.20 where the pressure
profiles of the experiments are compared. BT (I) sandstone did not offer any restriction
to generate shear rate and obtain emulsions.
71
Fig.4.19—Recovery production profile from CF1 and CF2
Fig. 4.20—Pressure profile from CF1 and CF2
72
Effect of permeability on the higher viscosity heavy oil
Oil recovery factors of the CF3 and CF4 are compared in Fig. 4.21 in order to
show the effect of permeability on the final recovery. The cores were saturated with
Kern River heavy oil of 3,500 cP. CF3 was performed with the BB (II) core sandstone
while CF4 was completed with the BT (II). Permeabilities of these cores are 152 mD and
2,015 mD respectively.
Emulsions were generated in CF3 at low injection rates, but due to high pressures
the experiment was terminated early. Fig. 4.22 shows that the permeability of the rock
plays a significant role in the emulsion generation. Low permeability restricts fluid flow
and increases shear rate which is favorable to emulsion generation. Despite the high oil
recoveries in CF4, emulsions were not generated.
73
Fig. 4.21—Recovery production profiles of CF3 and CF4
Fig. 4.22—Pressure profiles of CF3 and CF4
74
Effect of heavy oil viscosity on the final oil recovery
The effect of viscosity plays an important role on final recovery and therefore
should be carefully evaluated. Difference in viscosities between the low-viscosity
injected fluid and in-situ heavy oil may lead to by-pass of the heavy oil (viscous
fingering of drive fluid), reducing the final recoveries. Mai and Kantzas (2009)
concluded in their work that oil recovery decreases when oil viscosity increases.
Nanoparticle dispersion flooding; however, showed that the obtained oil
recoveries were mainly due to an IFT reduction as well as to the generation of emulsions
in those rocks with low permeability. These results are shown in the Fig. 4.23 and Fig.
4.24.
75
Fig. 4.23—Effect of oil viscosity in the CF1 and CF3
Fig. 4.24—Effect of oil viscosity in the CF2 and CF4
76
4.2.2 Effects of the ND2 on the in-situ emulsion generation and oil recovery
CF5 and CF6 were performed to investigate the effect of the ND2 in the oil
recovery. A reduction in the nanoparticle concentration would imply a reduction of the
elaboration cost of the dispersion. In addition, if the oil recovery is increased, the
economic results of the projects can be improved.
Core flooding 5: Louisiana heavy oil – Buff Berea (III)
The Core flooding 5 was carried out with a Buff Berea sandstone core saturated
with Louisiana heavy oil. Brine permeability of this core was 271 mD, with a 21.1%
porosity and an OOIP of 12.4 cm3. During the waterflooding stage, brine was injected at
1 ft/d. Oil recovery after waterflooding was 31.7% of the OOIP. Table 4.11 shows a
summary of the results of this experiment.
Table 4.11—Core flooding 5 experiment results
Core flooding 5
Heavy Oil type La HO
Nanoparticles [wt%] 2.0
NaCl [wt%] 0.5
Sandstone type BB (III)
Oil Recovered WF [ cm3] 3.9
RF after WF [%] 31.7
Oil Recovered NF [ cm3] 2.0
RF after NF [%] 16.2
Final RF [%] 47.9
77
ND2 was injected at different injection rates. Injection rates of 1 ft/d and 10 ft/d
did not produced additional oil. Emulsion and crude oil was produced with injection
rates higher than 100 ft/d. The oil recovery after ND2 flooding was 16.2 % of the OOIP.
Fig. 4.25 shows the oil recovery and injection profiles that resulted from this
experiment.
Fig. 4.25—Oil recovery and pressure profiles along the core flooding 5
78
Detailed results of the ND2 flooding are presented in the Table 4.12. It is
important to highlight that there was additional oil recovery due to the emulsion
generation at high injection rates (Fig. 4.26).
Table 4.12—Recovery factors and injection pressures at each injection rate, CF5
Injection
Rate
(ft/d)
Recovery
Factor
(%)
PV
Injected
Pressure
(psi)
o/w
Emulsions
1 0.0 0.3 34 no 10 0.0 1.7 72 no 100 4.0 4.2 185 yes 200 4.0 4.2 1,908 yes 300 8.1 6.3 2,455 yes
16.2 16.6
Fig. 4.26—Effluents collected from the CF5
79
Core flooding 6: Louisiana heavy oil – Bentheimer (III)
Core flooding 6 was carried out using a Bentheimer sandstone core saturated
with Louisiana heavy oil. Brine permeability of this core was 2,027 mD with a 24.1%
porosity and an OOIP of 14.9 cm3. During the waterflooding stage, 25.2% of the OOIP
was recovered. Results of the CF6 are shown in the Table 4.13.
Table 4.13—Core flooding 6 experiment results
Core flooding 6
Heavy Oil type La HO
Nanoparticles [wt%] 2.0
NaCl [wt%] 0.5
Sandstone type BT (III)
Oil Recovered WF [ cm3] 3.8
RF after WF [%] 25.2
Oil Recovered NF [ cm3] 3.5
RF after NF [%] 23.5
Final RF [%] 48.7
80
Fig. 4.27 shows the oil recovery and pressure profiles of the CF 6. The Oil
recovery profile exhibits an increment when the ND2 was injected at 100 ft/d, 200 ft/d
and 300 ft/d. However, effluents show that there was not stable emulsion generation; the
oil was produced primarily due to the IFT reduction and high flow rates (Fig. 4.28).
Fig. 4.27—Oil recovery and pressure profiles along the CF6. Due to the high rock
permeability, the pressure drop remains low when it is compared with the lower
rock permeability
81
Table 4.14 summarizes the results obtained after the ND2 was injected. Despite
the high rates, the injection pressure did not significantly increase. It means that the
injected fluid passed through the rock without any restriction hence it is believe that the
additional oil recovery was due only to the IFT reduction (Fig. 4.28).
Table 4.14—Recovery factors and injection pressures at each injection rate, CF6
Injection
Rate
(ft/d)
Recovery
Factor
(%)
PV
Injected
Pressure
(psi)
o/w
Emulsions
1 0.0 0.2 2.0 no 10 0.0 1.5 5.0 no 100 3.4 3.9 9.8 no 200 6.7 3.9 10.9 no 300 13.4 5.8 11.4 no
23.5 15.3
Fig. 4.28—Effluents collected from the CF6
82
Effect of the nanoparticle concentration on the emulsion generation and oil recovery
Fig. 4.29 and Fig. 4.30 intent to compare the effects on the oil recovery when
ND1 and ND2 are injected in two cores with the same permeability and saturated with
the same type of heavy oil.
Fig. 4.29 compared the oil recovery profiles from the CF1 and CF5; these
experiments were made with the Buff Berea sandstone core saturated with Louisiana
heavy oil. The ND1 was used in CF1 and ND2 in CF5. Additional oil was produced in
both core flooding as emulsions and crude oil; however, CF1 showed better recoveries.
Final recovery factor in CF1 was 64.3% while in CF5 was 48.7%.
Fig. 4.30 shows oil recovery profiles from CF2 and CF6. This set of experiments
was performed with the Bentheimer sandstone core saturated with Louisiana heavy oil.
CF2 was completed with the ND1 and CF6 with ND2. Oil was produced with the
nanoparticle dispersion flooding as crude oil. Emulsions were not observed in these
experiments. Final oil recoveries in CF2 and CF6 were 63.1% and 48.7 % respectively.
83
Fig. 4.29—Comparison of the ND1 and ND2 in Buff Berea sandstone
Fig. 4.30—Comparison of the ND1 and ND2 in Bentheimer sandstone
84
V. CONCLUSIONS AND RECOMMENDATIONS
5.1 Conclusions
Based on results from bulk emulsion generation and core flooding experiments
the following is concluded:
Stable heavy-oil-in-water emulsions can be created by mixing a silica
nanoparticle dispersion and heavy oil applying adequate shear rate through
mechanical means. Microscopic images and observed volumes in a graduated
cylinder indicated relative amount of emulsion present.
Emulsion generation was directly affected by nanoparticle concentration, NaCl
concentration, and the type of heavy oil.
Dispersion of 5.0 wt% nanoparticles generated the largest amount of emulsions.
The amount of emulsions was proportional to the nanoparticle concentration over
the attempted range of 0.5 – 5.0% wt% nanoparticle.
Adding Sodium Chloride to the nanoparticle dispersions had a positive effect in
the emulsion generations.
Pendant drop experiments demonstrated a reduction in IFT as the nanoparticle
and NaCl concentrations increased. We conclude that reduction of IFT assisted
with the creation of emulsions.
85
The viscosity of the emulsion presented a shear-thickening behavior. Viscosity
values of emulsions were significantly reduced compared with the original heavy
oil viscosity.
Core flooding experiments (5.0 wt% nanoparticles / 0.5 wt% NaCl) carried out
with the Buff Berea sandstone core, produced additional oil, both as o/w
emulsions and free crude oil. However, high injection pressures were observed as
a result of the low rock permeability; additionally, high injection rates were
required for incremental oil production. CF1, CF3 and CF4 recovered 32.3%,
20.2% and 24.2% respectively of the OOIP additionally to waterflooding (using
the nanoparticle solution).
Core flooding experiments conducted with the Bentheimer sandstone core,
produced additional oil; however, there was no observable (produced) emulsion
in these corefloods. It is possible that unstable emulsions were formed but broke
before exiting the core. Or incremental oil production was due to IFT reduction
without the necessity for emulsion formation. Nanoparticle dispersion flooding
recovered 39%, 64.1% and 30.2% of the OOIP in the experiments CF 2, CF 4
and CF 6 respectively.
The rock permeability was the most important parameter in stable in situ
emulsion generation. Effluents showed that emulsions were generated in the
lower permeability cores as a result of higher shear rates.
86
Although emulsions were not generated in high permeability cores, the
nanoparticle dispersion reduced the IFT, mobilizing the crude oil out of the core.
5.2 Future work and recommendations
Optimize emulsification of heavy oil by nanoparticles at lower shear than was
attained in this work in order to provide a system suitable for a field pilot.
o Investigate effects of heavy oil composition (saturates, aromatics, resins,
asphaltenes and total acid number) on emulsification using these types of
nanoparticles.
o Investigate changes in nanoparticle surface properties (vary degree of
hydrophilic modification) on emulsification of heavy oils of various
compositions.
o Conduct more core flooding experiments with different rock permeabilities
in order to determine the limits where the emulsions may be generated.
Due to the great amount of water necessary to obtain additional oil, analysis of
the effluent dispersions are required to evaluate that their properties remain
constant. If the properties do not change the dispersions may be recycled.
Emulsion breakers must be tested in order to find the one(s) which optimizes
breaking of produced emulsions.
87
REFERENCES
Adams, D.M. 1982. Experiences with Waterflooding Lloydminster Heavy-Oil
Reservoirs. Journal of Petroleum Technology 34 (8): 1643-1650. DOI:
10.2118/10196-pa
Alboudwarej, H., Felix, J., Taylor, S. et al. 2006. Highlighting Heavy Oil. Oilfield
Review 18 (2): 34-53.
Ali, S.M.F., Figueroa, J.M., Azuaje, E.A. et al. 1979. Recovery of Lloydminster and
Morichal Crudes by Caustic, Acid and Emulsion Floods. Journal of Canadian
Petroleum Technology 18 (1). DOI: 10.2118/79-01-02
Arditty, S., Schmitt, V., Giermanska-Kahn, J. et al. 2004. Materials Based on Solid-
Stabilized Emulsions. Journal of Colloid and Interface Science 275 (2): 659-664.
DOI: http://dx.doi.org/10.1016/j.jcis.2004.03.001
Arihara, N., Yoneyama, T., Akita, Y. et al. 1999. Oil Recovery Mechanisms of Alkali-
Surfactant-Polymer Flooding. Paper presented at the SPE Asia Pacific Oil and
Gas Conference and Exhibition, Jakarta, Indonesia. Society of Petroleum
Engineers 00054330. DOI: 10.2118/54330-ms.
Asghari, K. and Nakutnyy, P. 2008. Experimental Results of Polymer Flooding of Heavy
Oil Reservoirs. Paper presented at the Canadian International Petroleum
Conference, Calgary, Alberta. Petroleum Society of Canada PETSOC-2008-189.
DOI: 10.2118/2008-189.
88
Ashrafizadeh, S.N. and Kamran, M. 2010. Emulsification of Heavy Crude Oil in Water
for Pipeline Transportation. Journal of Petroleum Science and Engineering 71
(3–4): 205-211. DOI: http://dx.doi.org/10.1016/j.petrol.2010.02.005
Attanasi, E.D. and Meyer, R.F. 2010. Natural Bitumen and Extra-Heavy Oil. 2010
Survey of Energy Resources: 123-150.
Bragg, J.R. 1999. Oil Recovery Method Using an Emulsion. Patent: US 5855243. Exxon
Production Research Company Houston, Tx. United States.
Brook, G. and Kantzas, A. 1998. Evaluation of Non-Thermal Eor Techniques for Heavy
Oil Production. Paper presented at the Annual Technical Meeting, Calgary,
Alberta. Petroleum Society of Canada PETSOC-98-45. DOI: 10.2118/98-45.
Bryan, J. and Kantzas, A. 2009. Potential for Alkali-Surfactant Flooding in Heavy Oil
Reservoirs through Oil-in-Water Emulsification. Journal of Canadian Petroleum
Technology 48 (2): 37-46. DOI: 10.2118/09-02-37
Bryan, J.L. and Kantzas, A. 2008. Improved Recovery Potential in Mature Heavy Oil
Fields by Alkali-Surfactant Flooding. Paper presented at the International
Thermal Operations and Heavy Oil Symposium, Calgary, Alberta, Canada.
2008, SPE/PS/CHOA International Thermal Operations and Heavy Oil
Symposium SPE-117649-MS. DOI: 10.2118/117649-ms.
Butler, R.M. 1994. Steam-Assisted Gravity Drainage: Concept, Development,
Performance and Future. Journal of Canadian Petroleum Technology 33 (2).
DOI: 10.2118/94-02-05
89
Butler, R.M., McNab, G.S., and Lo, H.Y. 1981. Theoretical Studies on the Gravity
Drainage of Heavy Oil During in-Situ Steam Heating. The Canadian Journal of
Chemical Engineering 59 (4): 455-460. DOI: 10.1002/cjce.5450590407
Clark, B., Graves, W.G., Lopez-de-Cardenas, J.E. et al. 2007. Topic Paper #22 Heavy
Oil. National Petroleum Council.
Das, S.K. 1998. Vapex: An Efficient Process for the Recovery of Heavy Oil and
Bitumen. SPE Journal 3 (3): 232-237. DOI: 10.2118/50941-pa
Dusseault, M. 2002. Cold Heavy Oil Production with Sand in the Canadian Heavy Oil
Industry. Alberta Department of Energy.
Dusseault, M. and Baoci, X. 2011. Chops: Canadian Technology Finds Lucrative
Application Abroad. Energy Bridge: Canada's Oil & Gas Magazine.
Emadi, A., Sohrabi, M., Jamiolahmady, M. et al. 2011. Mechanistic Study of Improved
Heavy Oil Recovery by Co2-Foam Injection. Paper presented at the SPE
Enhanced Oil Recovery Conference, Kuala Lumpur, Malaysia. Society of
Petroleum Engineers SPE-143013-MS. DOI: 10.2118/143013-ms.
Falcone, G. 2009. Chapter 7 Heavy Oil Metering Applications. In Developments in
Petroleum Science, ed. Gioia Falcone, G.F.H. and Claudio, A.: Elsevier. Volume
54.
90
Fipke, S.R. and Celli, A.O. 2008. The Use of Multilateral Well Designs for Improved
Recovery in Heavy-Oil Reservoirs. Paper presented at the IADC/SPE Drilling
Conference, Orlando, Florida, USA. 2008, IADC/SPE Drilling Conference SPE-
112638-MS. DOI: 10.2118/112638-ms.
Fu, X., Lane, R.H., and Mamora, D.D. 2012. Water-in-Oil Emulsions: Flow in Porous
Media and Eor Potential. Paper presented at the SPE Canadian Unconventional
Resources Conference, Calgary, Alberta, Canada. Society of Petroleum
Engineers SPE-162633-MS. DOI: 10.2118/162633-ms.
Fu, X. and Mamora, D.D. 2010. Enhanced Oil Recovery of Viscous Oil by Injection of
Water-in-Oil Emulsions Made with Used Engine Oil. Paper presented at the SPE
Improved Oil Recovery Symposium, Tulsa, Oklahoma, USA. Society of
Petroleum Engineers SPE-129902-MS. DOI: 10.2118/129902-ms.
Gómez-Bueno, C.O., Marcano, N., and Sanchez, A.C. 1998. Orimulsion® a Fuel from
Heavy Oil, Engineered for Performance.
Greaves, M., Saghr, A., T. X. Xia, T. et al. 2001. Thai-New Air Injection Technology for
Heavy Oil Recovery and in Situ Upgrading. Journal of Canadian Petroleum
Technology 40 (3). DOI: 10.2118/01-03-03
Greaves, M. and Xia, T. 2001. Capri-Downhole Catalytic Process for Upgrading Heavy
Oil: Produced Oil Properties and Composition. In Canadian International
Petroleum Conference.
91
Head, I.M., Jones, D.M., and Larter, S.R. 2003. Biological Activity in the Deep
Subsurface and the Origin of Heavy Oil. Nature 426 (6964): 344-352.
Hincapie, R., Tovar, F.D., and Alvarez, C.E. 2011. Feasibility for the Application of in
Situ Combustion in Faja Petrolifera Del Orinoco (Fpo) Based in a Novel
Screening Criteria for the Technology. Paper presented at the SPE Enhanced Oil
Recovery Conference, Kuala Lumpur, Malaysia. Society of Petroleum Engineers
SPE-144027-MS. DOI: 10.2118/144027-ms.
Kaminsky, R.D., Wattenbarger, R.C., Lederhos, J. et al. 2010. Viscous Oil Recovery
Using Solids-Stabilized Emulsions. Paper presented at the SPE Annual Technical
Conference and Exhibition, Florence, Italy. Society of Petroleum Engineers
SPE-135284-MS. DOI: 10.2118/135284-ms.
Lu, X., Sun, S., and Xu, J. 2010. Application of Thermal Recovery and Waterflood to
Heavy and Extra-Heavy Oil Reservoirs: Analog Knowledge from More Than
120 Clastic Reservoirs. Paper presented at the International Oil and Gas
Conference and Exhibition in China, Beijing, China. Society of Petroleum
Engineers SPE-130758-MS. DOI: 10.2118/130758-ms.
Mai, A., Bryan, J., Goodarzi, N. et al. 2009. Insights into Non-Thermal Recovery of
Heavy Oil. Journal of Canadian Petroleum Technology 48 (3): 27-35. DOI:
10.2118/09-03-27
92
Mai, A. and Kantzas, A. 2009. Heavy Oil Waterflooding: Effects of Flow Rate and Oil
Viscosity. Journal of Canadian Petroleum Technology 48 (3): 42-51. DOI:
10.2118/09-03-42
Martínez-Palou, R., Mosqueira, M.d.L., Zapata-Rendón, B. et al. 2011. Transportation of
Heavy and Extra-Heavy Crude Oil by Pipeline: A Review. Journal of Petroleum
Science and Engineering 75 (3–4): 274-282. DOI:
http://dx.doi.org/10.1016/j.petrol.2010.11.020
Matthews, C.S. 1983. Steamflooding. Journal of Petroleum Technology 35 (3): 465-471.
DOI: 10.2118/9993-pa
McAuliffe, C.D. 1973. Crude-Oil-Water Emulsions to Improve Fluid Flow in an Oil
Reservoir. Journal of Petroleum Technology 25 (6): 721-726. DOI:
10.2118/4370-pa
Miller, K.A. 2005. State of the Art of Western Canadian Heavy Oil Water Flood
Technology. Paper presented at the Canadian International Petroleum
Conference, Calgary, Alberta. Petroleum Society of Canada PETSOC-2005-251.
DOI: 10.2118/2005-251.
Needham, R.B. and Doe, P.H. 1987. Polymer Flooding Review. Journal of Petroleum
Technology 39 (12): 1503-1507. DOI: 10.2118/17140-pa
Nelder, C. 2011. Heavy Oil of the Kern River Oil Field.
Oppelt, E.T. 2001. Environmental Impacts of the Use of Orimulsion. National Risk
Managment Research Laboratory.
93
Pickering, S.U. 1907. Emulsions. Journal of the Chemical Society 91: 2001-2021.
Pourabdollah, K. and Mokhtari, B. 2013. The Vapex Process, from Beginning up to
Date. Fuel 107 (0): 1-33. DOI: http://dx.doi.org/10.1016/j.fuel.2012.12.003
Revana, k. and Erdogan, M.H. 2007. Optimization of Cyclic Steam Stimulation under
Uncertainty. Paper presented at the Hydrocarbon Economics and Evaluation
Symposium, Dallas, Texas, U.S.A. Society of Petroleum Engineers SPE-
107949-MS. DOI: 10.2118/107949-ms.
Saigal, T., Dong, H., Matyjaszewski, K. et al. 2010. Pickering Emulsions Stabilized by
Nanoparticles with Thermally Responsive Grafted Polymer Brushes. Langmuir
26 (19): 15200-15209. DOI: 10.1021/la1027898
Sarathi, P.S. 1999. In-Situ Combustion Handbook -- Principles and Practices.
Selby, R., Alikhan, A.A., and Ali, S.M.F. 1989. Potential of Non-Thermal Methods for
Heavy Oil Recovery. Journal of Canadian Petroleum Technology 28 (4). DOI:
10.2118/89-04-02
Shah, A., Fishwick, R., Wood, J. et al. 2010. A Review of Novel Techniques for Heavy
Oil and Bitumen Extraction and Upgrading. Energy & Environmental Science 3
(6): 700-714.
94
Shell. 2012. Enhanced Oil Recovery. In http://www-
static.shell.com/content/dam/shell/static/future-energy/downloads/eor/eor-
brochure-2012.pdf, ed. Production, S.I.E.a. http://www-
static.shell.com/content/dam/shell/static/future-energy/downloads/eor/eor-
brochure-2012.pdf.
Shokri, A.R. and Babadagli, T. 2012. An Approach to Model Chops (Cold Heavy Oil
Production with Sand) and Post-Chops Applications. Paper presented at the SPE
Annual Technical Conference and Exhibition, San Antonio, Texas, USA.
Society of Petroleum Engineers SPE-159437-MS. DOI: 10.2118/159437-ms.
Shutang, G. and Qiang, G. 2010. Recent Progress and Evaluation of Asp Flooding for
Eor in Daqing Oil Field. Paper presented at the SPE EOR Conference at Oil &
Gas West Asia, Muscat, Oman. Society of Petroleum Engineers SPE-127714-
MS. DOI: 10.2118/127714-ms.
Sohrabi, M., Jamiolahmady, M., and Quraini, A.A. 2007. Heavy Oil Recovery by Liquid
Co2/Water Injection. Paper presented at the EUROPEC/EAGE Conference and
Exhibition, London, U.K. Society of Petroleum Engineers SPE-107163-MS.
DOI: 10.2118/107163-ms.
Thomas, S. 2008. Récupération Assistée Du Pétrole : Panorama. Oil & Gas Science and
Technology - Rev. IFP 63 (1): 9-19.
95
Turta, A.T., Chattopadhyay, S.K., Bhattacharya, R.N. et al. 2007. Current Status of
Commercial in Situ Combustion Projects Worldwide. Journal of Canadian
Petroleum Technology 46 (11). DOI: 10.2118/07-11-ge
Upreti, S.R., Lohi, A., Kapadia, R.A. et al. 2007. Vapor Extraction of Heavy Oil and
Bitumen: Review. Energy & Fuels 21 (3): 1562-1574. DOI:
10.1021/ef060341j
Xia, T.X. and Greaves, M. 2006. In Situ Upgrading of Athabasca Tar Sand Bitumen
Using Thai. Chemical Engineering Research and Design 84 (9): 856-864. DOI:
http://dx.doi.org/10.1205/cherd.04192
Xia, T.X., Greaves, M., Turta, A.T. et al. 2003. Thai—a ‘ hort -Distance Displacement’
in Situ Combustion Process for the Recovery and Upgrading of Heavy Oil.
Chemical Engineering Research and Design 81 (3): 295-304. DOI:
http://dx.doi.org/10.1205/02638760360596847
Zhang, T., Davidson, D., Bryant, S.L. et al. 2010. Nanoparticle-Stabilized Emulsions for
Applications in Enhanced Oil Recovery. Paper presented at the SPE Improved
Oil Recovery Symposium, Tulsa, Oklahoma, USA. Society of Petroleum
Engineers SPE-129885-MS. DOI: 10.2118/129885-ms.
Zhang, Y., Luo, P., and Huang, S.-S.S. 2010. Improved Heavy Oil Recovery by Co2
Injection Augmented with Chemicals. Paper presented at the International Oil
and Gas Conference and Exhibition in China, Beijing, China. Society of
Petroleum Engineers SPE-131368-MS. DOI: 10.2118/131368-ms.
96
Zhang, Y.P., Sayegh, S., and Huang, S. 2006. Enhanced Heavy Oil Recovery by
Immiscible Wag Injection. Paper presented at the Canadian International
Petroleum Conference, Calgary, Alberta. Petroleum Society of Canada
PETSOC-2006-014. DOI: 10.2118/2006-014.
Zheng, S. and Yang, D.T. 2012. Pressure Maintenance and Improving Oil Recovery Via
Water-Alternating-Co2 Process in Thin Heavy Oil Reservoirs. Paper presented at
the SPE Heavy Oil Conference Canada, Calgary, Alberta, Canada. Society of
Petroleum Engineers SPE-157719-MS. DOI: 10.2118/157719-ms.